MINISTRY OF PLANNING AND INVESTMENT PUBLIC EXPENDITURE TRACKING SURVEY REPORT

Transcription

1 MINISTRY OF PLANNING AND INVESTMENT PUBLIC EXPENDITURE TRACKING SURVEY REPORT (On the implementation of the Prime Minister s Decision 112/2007/QĐ-TTg of 20 July 2007: Policies to Support Services, Improve and Increase the Standard of Living, and Provide Legal Aid for Legal Awareness Raising)

2

3 PUBLIC EXPENDITURE TRACKING SURVEY REPORT (On the implementation of the Prime Minister s Decision 112/2007/QĐ-TTg of 20 July 2007: Policies to Support Services, Improve and Increase the Standard of Living, and Provide Legal Aid for Legal Awareness Raising)

4 Acknowledgements This Public Expenditure Tracking Survey (PETS) was undertaken in under the framework of the Country Programme of Cooperation between the Government of Viet Nam and UNICEF. This publication exemplifies the strong partnership between the Ministry of Planning and Investment, Dien Bien Province and UNICEF Viet Nam. The Survey was completed by a research team consisting of Ms. Nguyen Thi Kim Dung (Central Institute for Economic Management), Ms. Tran Nhu Trang (Center for Rural Progress) and Ms. Hoang Thi Thuy Nguyet (Academy of Finance) with technical guidance and support from Prof. Bernard Gauthier (University of Montreal). The Survey Report received valuable inputs and comments from experts of local departments and agencies in Dien Bien (Department of Planning and Investment, Department of Education and Training, Department of Finance, Provincial Statistics Office, Committee for Ethnic Minorities, agencies in Muong Cha and Dien Bien Dong districts) and ministries and committees at central level (Ministry of Planning and Investment, Ministry of Finance, Ministry of Education and Training, Committee for Ethnic Minorities). The Ministry of Planning and Investment and UNICEF Viet Nam would like to sincerely thank all those who contributed to this publication. iv

5 ABBREVIATIONS CIEM CEMA DEMA DDEMA DDOET DDOLISA DDPI DOET DOLISA DPI D112 Decision 112 MOF MPI PPC PETS PSO P135-II SEDP Central Institute for Economic Management Committee for Ethnic and Minority Affairs Department of Ethnic and Minority Affairs District Department of Ethnic and Minority Affairs District Department of Education and Training District Department of Labour, Invalids and Social Affairs District Division of Finance and Planning Department of Education and Training Department of Labour, Invalids and Social Affairs Department of Planning and Investment Ministry of Finance Ministry of Planning and Investment Provincial People s Commitee Public Expenditure Tracking Survey Provincial Statistic Office Program 135, phase II Socio-Economic Development Plan v

6 CONTENTS Executive Summary...x 1 Introduction...4 Motivation and objectives...5 Contents of the report Research methodology and tools Description of the program and context Objectives of the program Dien Bien Province and implementation of D Organizational arrangements for D The general budgeting process D.112 s rules and processes D.112 Resource allocation flows Main Findings Delays...25 The budget request preparation and submission process...27 The budget allocation process and the problem of delays...30 A significant delay in the transfer of funds from the schools to the households Measurement of potential fund leakage at various levels...38 Amount of funds transferred from the central government to Dien Bien Province...38 Amount of funds transferred from Dien Bien Province to the districts...38 Budget and funds released from the DPC to the DDOET...39 Release of funds from the district treasury to the schools...40 Release of funds from the schools to the households and fund leakage reported by the households 42 The poor practice of schools of not returning unused D.112 funds...46 Fund leakage in the form of beneficiaries paying back some expenses to the schools...48 Violation of general accounting procedures - a risk of fund leakage...48 vi

7 4.3 Equity issues...49 Issue of the coverage of D Issues of imprecise identification of beneficiaries Other shortcomings...51 Overly-complicated beneficiary selection procedures...51 Use of the support funds at the household level...52 User satisfaction and issue of limited access to information about the program...53 Issue of supervision and accountability...55 Poor collaboration between the main government stakeholders Concluding remarks and recommendations Concluding remarks Recommendations...60 Lessons learned for the future...64 References...65 Appendix...66 vii

8 TABLES Table 1. Main characteristics of the sampled districts...10 Table 2: Sampled communes in the two districts Table 3: Distribution of the household sample in the selected communes and districts Table 4. Main Characteristics of the sampled households...12 Table 5. Main Characteristics of the sampled schools...13 Table 6. Poverty rates in Dien Bien Province, (%)...19 Table 7. Time disparity between the official and the actual timeline for D.112 in Dien Bien Province (school years and )...25 Table 8. Frequency that schools verified the list of eligible beneficiaries with the key agencies Table 9. Actual budgeting processes for the four semesters of the and school years in the two surveyed districts...31 Table 10: Length of time for budget approval as reported by schools (number of days) according to vouchers/supporting documents and completed forms...32 Table 11. Time D.112 funds were transferred from the central government to Dien Bien Province Table 12. Time gap between the dates of cash withdrawal from the treasury and the dates of its disbursement to the students by the schools (number of days)...34 Table 13. Time lag in the budgeting procedure for D Table 14. Time resources were received by respondents in Muong Cha and Dien Bien Dong Districts for each semester of the school year...37 Table 15. Amount of funds transferred from the central government to Dien Bien Province for D Table 16. Funds released to Dien Bien Dong District over the two school years for all three levels of schools in the district...39 Table 17. Funds transferred in Dien Bien Dong District during both school years...39 Table 18. Funds transferred in Muong Cha District during both school years (in VND)...40 Table 19. Cash withdrawn, spent and carried forward...41 Table 20. Forms of delivery of D.112 support as perceived by the households and schools Table 21. Frequency of cash transfers among the surveyed schools ( )...43 Table 22. Time and average amount of resources received by respondents in Muong Cha and Dien Bien Dong Districts for each semester of the school year Table 23. Resource surpluses retained by the schools...46 Table 24. Number of schools withdrawing the support cash during the summer holiday...47 Table 25. Perceived failure in the extension of D.112 coverage...49 Table 26: Number of schools with documented dates of cash disbursement to students...56 viii

9 FIGURES Figure 1. Components of P.135 and D.112 and the scope of the study...18 Figure 2. Official organizational structure of D.112 in Dien Bien Province...20 Figure 3. Actual organization structure of D.112 in Dien Bien Province...20 Figure 4: Flow of information and funds for D Figure 5. Percentage of households that paid a commission for receipt of D.112 support...47 Figure 6. Percentage of households that reported having had to make a payment back to school for school expenses...48 Figure 7. Percentage of poor students who did not receive any support from D.112 although they regularly attended school, as reported by households...50 Figure 8. Perception of households and school managers on the seriousness of the issue of complicated procedures...51 Figure 9. Spending pattern of D.112 funds...53 Figure 10. Households satisfaction with D Figure 11. Percentage of recipients that had been informed about D.112 before each transfer Figure 12: Knowledge of respondents about D Figure 13. Reasons households were eligible for D ix

10 Executive Summary Over the past two decades, Viet Nam has achieved impressive economic growth and poverty reduction. However, there remains a large gap between urban and rural areas, and a high level of poverty persists, especially among ethnic minority groups. In 2008, the incidence of poverty among ethnic minorities was 50.3% compared to 9% for the majority group. Furthermore, despite increased public investment, the quality of education in minority areas remains poor and attendance remains low. In Dien Bien Province for instance, in the same year of 2008, only 88.4 out of 100 primary school-aged children attend primary school compared to the national average of 90.4 out of 100, and less girls attend school than boys 1. To respond to these development challenges and to ensure its commitments to the international treaties, particularly the Convention on the Rights of the Child (1989) and the Universal Declaration of Human Rights (1984) with specific articles on the rights to social security and protection, the Government of Viet Nam through the Ministry of Planning and Investment (MPI) adopted an integrated Socio-Economic Development Plan (SEDP) which included a number of programs that directly targeted access to public and social services by the poorest segments of the population. UNICEF Viet Nam has been working with MPI to enhance the social performance of SEDP through the use of a social audit approach, which provides information on the quality of social services and spending, with a focus on generating this information through citizen feedback. One of the tools, the Public Expenditure Tracking Survey (PETS), was used on a small-scale pilot basis in two districts in Dien Bien Province to assess the effectiveness of a cash transfer program for school children and to build capacity for the use of the social audit approach. PETS is a tool aimed at monitoring how much of public resources reach intended target groups and at diagnosing problems in service delivery so as to identify bottlenecks in program implementation and to generate information that will improve government decision-making process. It examines the manner, quantity and timing of release of resources to different levels of government, particularly the units responsible for the delivery of services and resources reaching beneficiaries. In Dien Bien Province, the pilot PETS was designed to collect quantitative information on the cash transfer program at all administrative levels, focusing at the service providers to the beneficiaries in order to get feedback on the challenges faced in delivering services. The cash transfer program analyzed in this study, D.112, has provided poor ethnic minority families a monthly allowance of 140,000 VND (about 7 USD) since 2007 with the aim of ensuring that school children do not drop out of school. The survey was implemented in December 2011 and January 2012 in two districts (out of nine) in Dien Bien Province (Dien Bien Dong District and Muong Cha District), and information was collected among program officials at all administrative levels as well as at all 65 primary and lower secondary schools in the two districts. 293 households (in which 604 students lived, 470 of which had received D.112 funds) were visited in eight communes to capture the perceptions and opinions of families on the quality and relevance of the services they were entitled to. The study monitored program implementation and resources transferred in and At the household level, the information collected concerned services provided during the school year. The key findings of the study reveals that D.112 beneficiaries, as well as local and school officials, expressed overall satisfaction with the program s assistance and goals. 77% of the surveyed households reported satisfaction with the program services- higher ratio of satisfaction is recorded in Dien Bien Dong District of 85% compared with 70% in Muong Cha District. Besides, more than 95% of the households reported that the support constituted a critical financial supplement that enabled improved education opportunities. One in ten households reported that their children would not have been able to attend school without the cash support. 1 UNICEF (2009), Analysis of the Situation of Children in Dien Bien, pg 80 x

11 Despite positive perception of the program among beneficiaries and local officials, various shortcomings were identified as follows: D.112 cash support was allocated either annually or bi-annually and there were few evidences of conditional checking on actual school attendance. This does not go in-line with the program s guidelines 2 which requires monthly cash transfer program for poor ethnic minority households conditional on actual school attendance. Besides, cash disbursements to beneficiaries were made several months late, long after the end of the school semester, and at some schools even during summer vacation or the following school year, further undermining the conditional nature of the transfer and reducing student attendance incentives. Delays in program disbursements were in large part explained by poor compliance with D.112 timeline regulations for budget preparation at almost all administrative levels. During the two school years of study, delays of three to 12 months were observed for D.112 budget preparation. Then significant delays were also observed in program implementation with the most serious delay in disbursement was observed at the stage of resource transfer from the central government to the provincial treasury. While official regulations stipulated that each fiscal year the eligible funds should be transferred to the province before 31 December, instead, for the second semester of resources allocated to the provincial treasury arrived five months late, and they were 11 months late for the first semester of Financial transfers from provincial to district treasuries were timelier, taking an average of two to eight weeks during the two years surveyed. Delays in cash transfers to beneficiaries were also observed at the school level. While two thirds of the schools reported delivering support to beneficiaries within ten days of cash retrieval from the district treasury, of the schools that reported financial data, about 10% took more than four months to deliver the cash transfers to the beneficiaries. Data from household survey with the hypothesis put forward by the team estimate that only 60-70% beneficiary received the support within one month of release; still about 20-30% only received the cash four to five months afterwards even. The average amount of the cash transfers reported by the beneficiaries was significantly lower than the official amount stipulated by the program s regulations. For the first semester of , beneficiaries in Dien Bien Dong District reported receiving 31% less program support than officially allocated (9% less in Muong Cha District ). For the second semester of , Dien Bien Dong District beneficiaries reported receiving 22% less than the official allocation (12% less in Muong Cha District). Payments of informal commissions were reported by the majority of beneficiaries in both districts. In addition, households also reported that program support was further reduced by school contribution when receiving cash transfers with averaged amount of about 140,000 VND 170,000 VND per student, i.e. more than one month of support per student. Resource leakage might have been exacerbated by the practice of school officials not returning program surpluses to the district treasury after transfers to beneficiaries. Despite clear budget regulations regarding surpluses, unused funds not distributed by schools to households were not returned to district treasuries and were accumulated in cash at the school level. For example, according to data reported by the DDOETs, the carried forward balance was more than one billion VND in Muong Cha District and about 700 million VND in Dien Bien Dong District during each semester in fiscal year No clear accounting evidence showing whether surpluses were returned to the treasury was obtained from the schools. Complex procedures for identifying, selecting and approving eligible children created a heavy burden for poor households and raised equity issues. In particular, the requirement for households to provide three notarized photocopies of their yellow card (poor card) to school officials constituted a major constraint to poor and remote households and possibly prevented some poor households from accessing program benefits. 2 According to circular 06/2007/TT-UBDT by CEMA, cash transfers to households must be made monthly at the end of each month based on the actual number of days children are present at school. xi

12 The issue of limited access to program support was raised by households. About 20% of the surveyed families reported that their children had not received the expected support despite meeting program eligibility criteria. The percentage of unsupported poor students among surveyed households reached more than 30% at some schools. The organizational structure of D.112 in Dien Bien does not fully follow program regulations for implementation, particularly the role of Department of Ethnic and Minority Affairs (DEMA) was quite limited in compared with their mandated responsibility of overseeing the implementation, monitoring and supervision of D.112. Accountability was limited due to a lack of standardized procedures for financial and program data management at the district, commune and school level. Data management quality was weak and varied across administrative entities. Poor data availability and accuracy were observed, particularly at the school level, through the inability of school managers to produce relevant records of financial transfers. For example, the lists of beneficiaries generally only provided names and signatures without dates or actual amounts transferred to the beneficiaries. Furthermore, data management procedures also varied across schools, communes and district authorities, which precluded direct consolidation or comparison of data obtained from these entities. More than six months after the end of D.112 (in May 2011) and contrary to regulations, final accounts of the program had not been completed. At the time of the survey in December 2011 and January 2012, despite regulations indicating that a final account settlement report for D.112 budget allocation had to be submitted to and approved by the People s Committee no later than 25 January of the following year, 3 these tasks had not been carried out at all levels in the province. Indeed, only one final account settlement for the spending in 2009 had been completed and none had been completed for 2010 or the first five months of Given these accounting lapses, the amounts reported at the district level and used in this study can only be considered a funds advance and hence it is difficult to make final conclusions on the usage of the funds and potential leakages as program accounts are not closed yet. One of the findings explaining for the contrary in delay and possible leakage of fund to the high rate of satisfaction was the fact that people were not well-informed about the program s objectives and timing and the amount of the transfers. Only about 40% of the respondents said they knew something about the program, and that information was very limited. After four years of implementation, such a low rate requires attention for other similar programs. Accordingly, households reported multiple uses of the support due to the fungible nature of the funds. 3 MOF Circular 108/2008/TT-BTC of 18 November 2008 guiding the closing of payment and the reporting of payment clearance at the end of the fiscal year. xii

13 Main Recommendations (1) Improve the existing public budget spending system at the local level It is highly recommended that Dien Bien authorities take appropriate measures to deepen understanding of the inefficiencies and limitations in the execution of D.112 to move forward to a more a transparent and easy public budget administration system. Such important indicators as precise identification of beneficiaries, timeliness and volume of support delivery, and improvement in achievement of the program s outcomes should be seriously taken into account. This includes more reforms regarding budget preparation and submission which may start from the procedure for the identification, selection and approval of eligible children. Checking and verification by commune authorities might be sufficient provided parallel measures are implemented to ensure accountability. (2) Take decisive action to stop all of the potential fund leakages and losses. In order to ensure that the program s limited resources reach the targeted beneficiaries without any leakage, a number of decisive actions should be taken such as immediately halting the practice of making the beneficiaries sign receipts before being paid; prohibiting the practice of paying some service fee; having more supervision and control over the practice of deducting part of the cash support to cover school expenses ; and strengthening compliance with government regulations concerning the return of inactive funds by schools. Last but not least, more awareness raising and public participation is needed to enhance knowledge among households of the objectives, rules and operational procedures of the program in order to reduce the risk of leakage of limited funds. Possible channels could be parents associations, commune officers, and village heads, and they would be capable of overseeing the whole process. In order to draw a final conclusion, final accounts settlement for the spending each semester should be carried out at all levels, pursuant to the State Budget Law. It is also recommended that program regulations clearly indicate the local authorities and schools duty to report publicly and widely the nature of program activities by means of quarterly and annual reports, announcing the names of the beneficiaries, and the amount and dates of the financial support for each transfer. Reports by independent auditors should also be made available upon request. (3) Ensure transparency during program implementation. Increased access to information should be provided to the beneficiaries. During the course of implementation, the schools and local authorities should enact more measures to raise awareness of the program among the beneficiaries in order to ensure greater transparency and more effective use of the support. (4) Enhance public participation. Enhanced public participation could be achieved through formalizing participation and consultation activities during the budgeting process, particularly the transfer steps. (5) Improve supervision and accountability. For the transparency of the program as a whole, progress at each stage of the program should be reported with specific performance indicators. Budget allocation and budget allocation norms should 1

14 be published via appropriate forms at all levels, especially at the frontline level for service providers and service users. The filing system and records at all management levels should be improved. This is particularly important at the household level. The DEMA and the DDEMA should be assigned as the appropriate agencies to monitor the program in terms of budget allocation, disbursement mechanisms, and payment to beneficiaries. School boards and citizen inspection committees at the commune level could be mobilized to monitor payment and utilization of the provided resources. An organization independent from the program s executive management, the program-related government agencies and the project implementers should be hired to carry out an evaluation and prepare a relevant assessment report. (6) Improve the management structure and coordination Among the various issues faced by D.112, the management structure, capacity and coordination deserves serious attention as the credibility of programs like D.112, as well as its long-term sustainability, crucially depends on these things. (7) There are still certain issues encountered by this research but have not been explored sufficiently. They suggest further study for valuable policy implications such as: Impacts on the studying quality of the pupils who receive the support; Repetition of the study on the same group who will continue to receive support from Decree 49 so as to compare and collect policy implications on the scheme of implementation and levels of support. 2

15 3

16 CHAPTER 1 4

17 Introduction Motivation and objectives Over the last decades, Viet Nam has made remarkable improvements in enhancing income and poverty alleviation. Poverty and the incidence of hunger in the country has declined significantly. Viet Nam has well exceeded the millennium development goal (MDG) of halving the proportion of people whose income is less than one dollar a day between 1990 and However, there is a large gap between urban and rural areas, and a high incidence of poverty in mountainous areas, especially among ethnic minority groups. The pace of poverty reduction among minority groups is slower than that of the majority groups (the Kinh and the Chinese). In 2008 the incidence of poverty among ethnic minority people was 50.3 %, much higher than the rate among the Kinh and the Chinese (nine %). Accounting for one eighth of the total population, ethnic minority people make up over 50 % of the poor, especially in terms of food poverty (MPI, 2008). Harsh natural conditions, the weather, limited access to major social services, and lack of market access are among the factors that have negatively impacted ethnic minorities living in mountainous and geographically remote communes and villages. To respond to these evolving development challenges and to ensure its commitments to the international treaties, particularly the Convention on the Rights of the Child (1989) and the Universal Declaration of Human Rights (1984) with specific articles on the rights to social security and protection, the Government of Viet Nam adopted an integrated Socio-Economic Development Plan (SEDP) to address economic, social, and environmental sectors on a regular basis. The SEDP is the Government s primary national planning tool, focusing on the economic sector, the social sector, the environment, and national defence and security. The Government also launched a broad number of supporting programs that directly targeted the most serious pockets of poverty throughout the region, and this support improved access to public and social services among poorer segments of the population. UNICEF Viet Nam has been working with the MPI to enhance the social performance of the SEDP through the use of the social audit approach, which provides information on the quality of social services and spending, with a focus on generating this information through citizen feedback. One of the tools, the Public Expenditure Tracking Survey (PETS), was used on a small-scale pilot basis in two districts in Dien Bien Province to assess the effectiveness of a central government cash transfer scheme for school children and to build up local capacity for the use of the social audit approach for the target groups including DPI and related departments of the province such as DOET, DOLISA, DEMA, PSO and their subsidiaries at district level. The study tracks the flow of social assistance from the province to the districts to the schools and then to eligible students at primary and lower secondary schools. PETS is a tool aimed at monitoring the amount of public resources that reach the intended target groups and at diagnosing problems in service delivery. It seeks to identify bottlenecks in program 5

18 implementation and to generate information that will improve the Government s decision-making process. It examines the manner, quantity and timing of release of resources to different levels of government, particularly the units responsible for the delivery of services and resources and services reaching beneficiaries. In Dien Bien Province, a pilot PETS was designed to collect quantitative information on the cash transfer program at all administrative levels and from the service providers and beneficiaries in order to get feedback on the challenges faced in delivering services. The cash transfer program analyzed in this study, D.112, is part of National Target Program P.135 Socio-Economic Development Program for Extremely Difficulty Communes in Ethnic Minority and Mountainous Areas. It has provided poor ethnic minority families a monthly allowance of 140,000 VND (about 7 USD) since 2007 with the aim of ensuring that school children do not drop out of school. The survey was implemented in December 2011 and January 2012 in two districts (out of nine) in Dien Bien Province (Dien Bien Dong District and Muong Cha District), and information was collected among program officials at all administrative levels as well as at all 65 primary and lower secondary schools in the two districts. 293 households (in which 604 students lived, 470 of which had received D.112 funds) were visited in eight communes to capture the perceptions and opinions of families on the quality and relevance of the services they were entitled to. Contents of the report The report is comprised of six sections: Section 1 contains the introduction; Section 2 presents the survey methodology and samples; Section 3 discusses the cash transfer program and the context of the survey; Section 4 analyses the organizational structure and rules of the program; Section 5 presents the main findings of the study regarding the program s bottlenecks, sources of delay, and potential leakage and equity issues; and Section 6 presents the conclusions and recommendations. 6

19 7

20 CHAPTER 2 8

21 Research methodology and tools This pilot PETS study was realized in three main steps, including a desk review of available documents, an institutional analysis and rapid data assessment phase, and a field-level survey pilot. Desk review The first step, a desk review, was conducted to collect and analyze the available information and data. The review covered a large range of documents including official documents on P.135 and D.112, 4 and all the official decisions on the establishment of the steering committee for D.112 at the central and provincial level. Other sets of documents reviewed were comprised of related administrative, financial, inspection and auditing reports. A review of the documents provided input for the institutional reports and helped identify the key stakeholders and resource flows within D.112. Institutional analysis and rapid data assessment An institutional analysis report was produced that examined the administrative structure and allocation rules within the program. This step included a rapid data assessment of the information available, which consisted of interviews with all related departments at the provincial, district and commune level that were involved in D.112 planning and implementation. Interviews were conducted in October and December 2011 with provincial and district representatives of the DEMA, the DOF, the DOET, the DOLISA, the DPI, the PPC, commune Peoples Committees, heads of villages/hamlets, and school management, as well as the beneficiaries. Questionnaire templates were used for the interviews with the various respondents. Field survey Survey questionnaires for all administrative levels, schools and beneficiaries were drafted based on information collected during the institutional analysis and rapid data assessment phase. The questionnaires were pre-tested at various levels including at six schools and 15 households in two 4 Decision 07/2006/QD-TTg of 10 January 2006, Decision 112/2007/QD-TTg of 20 July 2007 and Decision 101/2009-QD-TTg, respectively. 9

22 communes. Following the pre-test, the questionnaires were revised and the survey implementation guidelines for enumerators were finalized. Data collection was implemented in December 2011 and January Sampling Two districts were selected for data collection, Muong Cha District and Dien Bien Dong District. One of the main criteria used for district selection was the presence of a high percentage of poor ethnic minority households (around 50% in each district) and hence a relatively high number of child beneficiaries of D.112 (the two districts accounted for about 30% of the total beneficiaries in Dien Bien Province). Geographically, the two selected districts are located not far away from the provincial capital city, Dien Bien. They share many similar socio-economic characteristics, and at the same time they are significantly different in terms of access to infrastructure. Muong Cha District is flat and has a better asphalt road network, and Dien Bien Dong District is a mountainous area and the roads are much worse (see Table 1). Table 1. Main characteristics of the sampled districts Indicators Muong Cha District Dien Bien Dong District Dien Bien Province Population 52,655 56, ,000 % of ethnic minorities Poverty rate (2009) Number of communes in P.135 (% in the province) 95% 99% 78% 55.1% 49.8% 33.6% 9 (15%) 13(22%) 59 Geography Number of schools 45 km from Dien Bien City, flater, better asphalt road network 60 km from Dien Bien City, more terrain, lower quality roads Primary Lower secondary (% in the province) (% in the province) 21(12%) 23(13%) (16%) 15(13%) 115 Number of D.112 beneficiaries ( ) Primary school Students 4,552 (18%) 3,358 (14%) 24,736 (100%) (% in the province) 4,552 (18%) 3,358 (14%) 24,736 (100%) 12,915 (100%) Lower secondary school Students (% in the province) 2,109 (15%) 1,576 (12%) 12,915 (100%) Source: Authors calculation based on local government documents. In each of these two districts, 300 households in four communes were selected for data collection. 5 5 The selection of communes was made by the Dien Bien Provincial PMU using options randomly computed by the study team for each district to ensure the presence of both upland and lowland beneficiaries. 10

23 Household selection was based on the benficiaries attendance at school. From each of the 24 primary and lower secondary schools located in the selected communes, 10 to 15 beneficiaries (in proportion to the number of beneficiaries) were identified using the list of D.112 beneficiaries during the second semester of the school year. Table 2: Sampled communes in the two districts Muong Cha District Muong Luan Na Son Pu Nhi Luan Gioi Dien Bien Dong District Huoi Leng Si Pa Thin Muong Muon Muong Tung In addition, a school-level questionnaire was administered at all 65 primary and secondary schools in the P.135 communes in the two selected districts 6. Table 3: Distribution of the household sample in the selected communes and districts Districts Sampled communes Sampled districts Muong Cha Dien Bien Dong Muong Cha Dien Bien Dong Dien Bien Province Number of D.112 beneficiaries ( ) Number of surveyed Households Number of surveyed students receiving D.112 support % of surveyed beneficiaries vs. total D.112 beneficiaries 1,940 1,410 6,661 4,934 37, % 14.1% 4.1% 4.0% 1.2% Source: Dien Bien Pilot PETS 2012 Data was collected at 293 households in which 787 children (aged six to 21) lived, 604 of whom attended school. Among those students, 553 attended the 24 sampled schools and 470 had received D.112 support in , representing 14 % of the total number of beneficiaries from the sampled communes in the two districts. Households characteristics Table 4 presents the main characteristics of the household part of the sample (see Table A1 in the appendix for more details). 6 The school questionnaire was composed of a data form sent to school officials and a school questionnaire administered by enumerators to school directors during a school visit. 11

24 The majority of the respondents in Muong Cha District were ethnic Hmong (63%), and 30% were ethnic Thai. In Dien Bien Dong District, the ethnic Thai were the majority (65%) and the Hmong represented 24% of the sample. Among the respondents, only about 10% had completed primary school and the proportion was about the same for their spouses. All the interviewed households had a yellow card in 2011, but only about 82% did in The average number of children in each family in Muong Cha District was slightly higher than in Dien Bien Dong District (2.87 versus 2.42), yet the ratio of children attending school full-time in Muong Cha District was slightly lower, 73% versus 79% in Dien Bien Dong District. The average annual household income was similar in both districts, around 13 million VND. Table 4. Main Characteristics of the sampled households Muong Cha District Dien Bien Dong District Total Number of sampled households Male (%) Ethnic groups Thai (%) Hmong ( %) Other (%) Completed first grade or never went to school (%) Completed grade 1-5 (%) Completed grade 6-12 (%) Married (%) Average number of children (aged 6-21) Children attending school (%) Yellow (poor) card status (2011) (%) Yellow (poor) card status (2010) (%) Average annual income (mil. VND) Source: Dien Bien Pilot PETS School characteristics A total of 65 questionnaires from school management were received from the two districts, including 28 from Muong Cha District (17 primary and 10 lower secondary schools, and one school that was both a primary and a lower secondary school) and 37 from Dien Bien Dong District (22 primary and 15 lower secondary schools) (see Table 5 and Table A2 in the Annex for more details). Each school had on average 17 classrooms, and about half of those had access to electricity. Most of the schools had separate toilets for males and females and most had a landline telephone. 20% of the schools surveyed did not have access to the Internet. Most of the schools were located near the commune center, from which it took a long time to reach the district center. The estimated average 12

25 time for school managers to travel to the district center was about one and a half hours and the longest commute was three hours. The schools across the two districts were staffed in a fairly similar manner: each had about 35 staff members and about 80% were teachers. The schools in Muong Cha District had on average more pupils than those in Dien Bien Dong Dostrict (300 and 269 respectively). The percentage of minority students in Dien Bien Dong District was larger than in Muong Cha District (95% and 90% respectively). However, the ratio of D.112 beneficiaries in Muong Cha District was significantly greater than in Dien Bien Dong District, almost 60% versus 46 %, and the percentage of those who were minorities was also larger, 63 % versus 47 %. Table 5. Main Characteristics of the sampled schools Muong Cha District Dien Bien Dong District Total Number of schools sampled Schools receiving boarding pupils (%) 53% 62% 58% Separate toilets for males and females (%) 93% 91% 92% Access to a telephone (%) 93% 95% 94% Internet connection (%) 79% 81% 80% Travel time by motorbike from school to (minutes) Health facility or hospital Commune center District treasury District DOET Average number of staff members Average number of teachers Average number of students ( ) Average number of minority students ( ) Percentage of minority students 90% 95% 94% Percentage of D.112 beneficiaries 59% 46% 52% Percentage of minority beneficiaries among D.112 beneficiaries 63% 47% 54% Source: Dien Bien Pilot PETS

26 CHAPTER 3 14

27 Description of the program and context 3.1 Objectives of the program In 1998, the Government of Viet Nam launched a National Target Program entitled Socio-economic Development Program for Communes with Extreme Difficulties in the Mountainous and Ethnic Minority Areas (P.135) in order to support ethnic minority groups. Initially, the program was to be implemented over a seven-year period from 1998 to 2005 (P.135-I), and it was then extended for a second phase (P.135-II) from 2007 to The objectives of P.135 were to promote market-oriented agriculture and economic development, and to increase access to basic services and infrastructure in order ultimately to eradicate hunger and reduce poverty in the targeted communes and villages. Implementation of P135-I involved the participation of 2,410 of Viet Nam s poorest communes and villages, covering around 1.1 million households and six million individuals. A total of 8,434 billion VND (about 600 million USD) was disbursed from the central government budget and 240 billion VND from local government budgets to support the construction of more than 20,000 small infrastructure projects at the inter-commune and commune levels. P.135-II targeted 1,848 poor and ethnic communes and 3,374 villages, with an estimated funding of about 7,000 billion VND (450 million USD). The program consisted of four main components: Production development and economic restructuring: The provision of seed; livestock and husbandry; fishery breeding; fertilizer and insecticides; the extension of new models of agricultural development; the provision of machinery, equipment, and production tools. Local infrastructure development: The construction of essential local infrastructure such as inter-village roads, small-scale irrigation works, rural schools, clinics, collective water supply stations, grid electricity, community meeting halls, and local markets. Capacity building: The provision of training courses to the commune, village or hamlet leaders, local staff and others on project management, agriculture techniques and technology; and developing household economy. Improvement of social and cultural livelihoods: Supporting students from poor households who attended boarding or semi-boarding schools; sponsoring cultural development (e.g. traditional culture events); and providing legal assistance to the poor. 15

28 P.135-II provided financial support to 926,326 children and students from poor households who attended kindergartens and schools. It was perceived as positively matching the expectations and aspirations of the Vietnamese people during inspections and assessment missions by the National Assembly and it is expected to be extended to a third phase in the coming years. Decision 112 From the beginning, great importance was attached to the fourth component of P.135 II, Improvement of social and cultural livelihoods. The improvement of socio-cultural livelihoods was achieved by improving access to education, healthcare, clean water, electricity and other key services for people in poor households. In terms of education, great attention was drawn to the increased rate of school enrollment and attendance by children from the P.135-II target households. However, there were marked gaps between gross and net enrollment rates at all levels in the P135-II communes compared to the national levels. For instance, the gross and net enrollment rates at the primary schools in the P135-II communes were 85% and 78%, respectively while the national averages were 104% and 89%. Gaps of more than 20 percentage points were also reported at secondary schools. 7 In Dien Bien Province in particular, school attendance was lower than the national average. For instance in 2008 the school attendance rate of children aged six to nine years old was 88.4% compared to 90.4% at the national level. The proportion of children who never attended school (10.4%) was also higher than the national average (8.1%) 8. In order to narrow this gap, D.112 was signed by the Prime Minister in July The main objective was to provide direct support to improve access to key public services in the targeted communes. D.112 consisted of four main components, including direct support through a cash transfer program to children from poor households who were studying at state schools. Initially planned for the period , the program was extended to May 2011, the last month of the academic year. 9 Nationwide, D.112 covered a large number of beneficiaries, ranging from pre-school students to adults in poor households in P.135-II communes and villages. By the end of 2009, 1, billion VND had been disbursed by the central government, supporting 926,326 kindergartners and students at semi-boarding schools In addition, there were considerable gaps between the actual enrollment rates and the targeted level for 2010: a ten percentage point shortfall was reported between the actual and targeted level at the primary school level, and gaps as high as 12 percentage points were found at the secondary level (UNDP (2008)) 8 UNICEF (2009), Analysis of the Situation of Children in Dien Bien, pg 80 9 The decision went into effect on 5 September 2007 and initially targeted only students at semi-boarding schools. It was amended in 2009 (Decision 101) to cover all students from poor households studying at state schools in the area. 10 Draft of Project Document of Program 135, phase 3 (8th draft) 16

29 Nguồn: PETS Dien Bien, photo taken in a surveyed school of Dien Bien Dong district Scope of the study This pilot PETS was limited to the investigation of the first component of D.112, i.e. the support for children from poor households attending primary or lower secondary schools. The study did not encompass support for students attending upper secondary schools, which were managed by the DOET through a different management scheme. Figure 1 illustrates the components of P.135 and D.112 and the scope of the study. 17

30 Figure 1. Components of P.135 and D.112 and the scope of the study P135-II Component 1: Production support Component 2: Infrastructure Component 3: Capacity Building Component 4: Improvement of people s livelihood & access to social services D Education support for children from poor households attending school 2. Improved sanitation 3. Support for culture and communication 4. Enhanced access to legal services 3.2. Dien Bien Province and implementation of D.112 Dien Bien Province, located in the mountainous northwest region, is one of the poorest provinces in Viet Nam. The total population in 2009 was people and is comprised of 19 ethnic minority groups, including approximately 37.99% ethnic Thai, 34.8% Hmong, 18.42% Kinh (Viet) and a variety of other groups such as Dao, Khmu and Ha Nhi. The province covers an area of 9,563 square kilometers and shares a 360 km long border with Laos and a 40.8 km border with China. Administratively, Dien Bien is divided into nine districts and towns11 with 112 communes (in rural areas) and wards (in urban areas). Over the last few years, the province s economy has grown steadily due especially to important government investment. Local public spending on education increased considerably, from 15% in 2005 to 28.3% in 2008, and this money was used mainly for infrastructure development. In terms of absolute figures, it is even more significant that the amount of public spending for 2008 was 763 billion VND, almost triple of the amount in 2005 (253 billion VND). The standard of living has, therefore, gradually improved. Ha Noi Dien Bien Hue However, social services continue to be of concern, and the province remains ranked among the poorest, with more than 33% of the households living below the poverty line in 2009 (see Table 6). A large majority of them (96.7%) were from ethnic minority groups and the province was one of the largest recipients of D.112 benefits Dien Bien Phu (the provincial capital), Muong Lay Town and seven rural districts: Dien Bien, Dien Bien Dong, Muong Cha, Muong Nhe, Tua Chua, Tuan Giao and Muong Ang.

31 Table 6. Poverty rates in Dien Bien Province, (%) Administrative units Percentage of the poor households Dien Bien Province Dien Bien Phu (the provincial capital) Muong Lay District Muong Nhe District Muong Cha District* Tua Chua District Tuan Giao District Dien Bien District Dien Bien Dong District* Muong Ang District Source: Dien Bien DPI, Note: The two sampled districts are marked with (*). 3.3 Organizational arrangements for D.112 According to Decision 112/2007QD-TTg, at the national level the Committee for Ethnic Minorities Affairs (CEMA) was the key government agency responsible for the execution and implementation of D.112, with funds coming mainly from P.135-II. Its tasks included making annual budget estimations; integrating them into the P.135-II master plan; undertaking monitoring and evaluation; and reporting periodically about D.112 to the Prime Minister. The role of coordinator was assigned to the MPI and the Ministry of Finance, and they helped the CEMA to estimate and allocate the annual budget. At the provincial level, the Provincial People s Committee (PPC) was assigned similar responsibilities, adapting and implementing central policies at the local level with assistance from the provincial Department of Ethnic Minorities Affairs (DEMA). The PPCs had discretion to specify and adapt the central regulations on organizational structuring of D.112 to the local conditions. 12 Furthermore, according to memorandum 588/UBND-NN of Dien Bien PPC, the DEMA was tasked with providing guidance to the provincial districts and cities as they made a budget plan for submission to the PPC. The DEMA was also formally responsible for providing technical instructions, performance monitoring and evaluation of all four components of D.112 implementation. Figure 2 illustrates the official organizational structure of D.112 as specified by Decision 112/2007/ QD-TTg. 12 Article 5, clause b of Decision 112/2007/QD-TTg. 19

32 Figure 2. Official organizational structure of D.112 in Dien Bien Province GOV The Prime Minister Central level: MPI Central level: CEMA Central level: MOF Sở KH-ĐT Provincial People's Committee => DEMA Provincial DOF Commune level Schools Coordination: Direction & Guidance: Beneficiary Households Source: Authors, adapted from Decision 112/2007/QD-TTg. Although the DEMA was officially responsibile, in practice it was the Dien Bien DOET that played the role of focal point during the implementation of D.112. The DOET was in charge of providing direction to the District Divisions of Education and Training at the district level (DDOETs) and to upper secondary schools in preparing D.112 annual budget plans; identifying potential beneficiaries; and monitoring, evaluating and reporting on the progress and performance of the program to the PPC. The DOET s central role and the DEMA s secondary role in D.112 implementation are illustrated in Figure 3. Figure 3. Actual organization structure of D.112 in Dien Bien Province The Prime Minister Central level MPI Central level MoET Central level: MOF Central level: CEMA Provincial level- DPI UBND => (Sở GD&ĐT) DOF DEMA District level - DDOET DDEMA Schools Households... Households Subordination Coordination: Source: Authors, adapted from the Dien Bien PPC s memo 588/UBND-NN of 16 June

33 The main explanation for the difference between the official regulations and the actual roles played by the provincial departments was the perceived lower capacity of the DEMA as an implementing agency The general budgeting process The general budgeting process The general budgeting process starts with the fiscal year, which in Viet Nam corresponds to the calendar year (January 1 to December 31). Annually, the budget preparation guidelines for the next year for all public expenditure programs (such as D.112) are initiated by the central government. Prior to 31 May the Prime Minister issues a Directive guiding preparation of the annual socio-economic development plan and the annual budget. Based on the Directive, the MOF and the MPI issue circulars providing detailed guidance on budget preparation to the central ministries, government agencies, and provinces. The same process is followed from the provincial to the district level, and from the districts to the schools. In contrast, the process of determining budget needs, consolidation and submission moves in an upward direction: Lower administrative levels prepare and consolidate budget requests and submit them to higher levels. For example, within D.112 the target 135 communes (including communes with 135 villages) submit budget needs to districts, districts consolidate all the needs of their communes for submission to the province, and the provinces submit requests to the central level. The State Budget Law does not specify the dates for budget submission at the local level (i.e. the provincial, district and commune level). It delegates this responsibility to the Provincial People s Committee (PPC) and only requires submission of the budget to the national government prior to 20 July. At the central level, the MPI and the MOF are responsible for the review and consolidation of local budget plans into the national budget plan. They then submit it to the Prime Minister. After reviewing this plan, the Prime Minister submits it to the National Assembly (NA) for final review and approval at its second annual session, which is usually held in November. The process of budget decisions and allocation starts at the central level, where the National Assembly decides on the budget appropriation for the local level prior to 15 November. At the local level, upon the consolidation of plans of all sectors and districts, DOF submitted to Province People s Committee for present and consultation to the Provincial People s Council (PPC) for decision on the budget apportionment for the districts prior to 10 December. The District People s Council (DPC) determines the budget of communes and other expenditure programs (such as D.112 for schools). All of these must be completed before 31 December of the current year. It is from this allocation that local spending agencies, e.g. commune People s Committees and schools, withdraw cash from the district state treasury and make monthly transfers to students the next year. Table A3 in the appendix illustrates the general budgeting process. 13 It was reported by the DEMA during the first and second meetings with the study team that the only activity they had carried out for D.112 was to receive the list of beneficiaries sent to them by the DOET and to participate in (only) one inspection mission organized by the PPC. 21

34 3.5. D.112 s rules and processes The budget preparation and spending for P.135 and its D.112 sub-component had to conform to the general state budget management regulations as well as to D.112 s rules. 14 In particular, beneficiaries of D.112 transfers had to meet various eligibility criteria: 15 be a member of a poor household (The criteria for determining whether a household is poor is formally defined in Decision 170/2005/QD-TTg. Physical evidence of the households poverty status is a yellow card issued by local authorities); or be part of a household living in an extraordinarily poor commune, village, or hamlet in region II, which were ranked as targets of P.135-II; attend school full-time (kindergarten, primary, lower or upper secondary school); and be a semi-borderer, staying near or at the school. 16 The last requirement was removed in 2009 and coverage was extended to all poor students in the targeted regions. 17 Accordingly, starting in the school year, all the poor students in the P.135 communes or villages attending school full-time were eligible for support. Officially, the duration of support corresponded to the actual time spent at school, but no longer than nine months per calendar year. Program support could take two forms: (a) direct cash transfer or (b) learning aids and school meals. Decision on specific choices had to be jointly made by the children s parents (or their representatives) and school leaders during an annual school meeting. Furthermore, cash transfers had to be made monthly at the end of each month based on the actual number of days students were present at school D.112 Resource allocation flows According to the state budget regulations, D.112 funds remained within the state treasury system through credit and debit functions during the allocation process and it was only when the budgets reached the schools that funds were cashed out and distributed to the targeted families. Upon receiving budget notification (from the PPC, the DOF and the DOET and then from the DPC, the DDOF and the DDOET to the schools), recipient schools withdrew cash from the district treasury and made disbursements to the target households based (officially) on the children s attendance. Figure 4 describes the flows of information and resource allocation toward households. 14 According to Decision 210/2006/QD-TTg, capital allocation to a specific NTP was done based on its own allocation principles, norms and criteria. The norms and criteria in turn were determined based on the program s objectives, target, size, organizational structure and other program-specified characteristics. Such capital allocation to the NTPs ensured both the objectives of development and the provision of resources to disadvantaged areas, including mountainous, border, island, and ethnic minority areas, to narrow the economic development gap and enhance the standard of living of the local population. 15 Circular 06/2007/TT-UBDT by the CEMA of September 20, Decision 112/2007/QD-TTg, of July Decision 101/2009/QĐ-TTg of 5 August Circular 06/2007/TT-UBDT by the CEMA. 22

35 Figure 4: Flow of information and funds for D.112 State Treasury (1) MOF (2b) Provincial State Treasury (3b) (2a) (3a) Provincial DOF DDOF District State Treasury (4) DDOETs (5) (6) Primary and lower secondary schools (7) Households Cash flow Information flow Source: PETS study team, based on program regulations. The relationships described in the figure are as follows: (1) The Ministry of Finance (MOF) informed the State Treasury of the amount of central budget transfers to the DOF. (2a) The State Treasury transferred the central budgets to the Provincial Department of Finance (PDOF). (2b) The State Treasury informed the Provincial State Treasury of the amount of the central budget transfers to the DOF. (3a) The Provincial State Treasury transferred the funds to the District Division of Finance and Planning Division (DDOFP). (3b) The Provincial State Treasury informed the District State Treasury of the amount of the funds transferred to the DDOFP. (4) The DDOFP informed the DDOET of program budget estimates for primary and secondary schools. (5) The DDOET informed primary and lower secondary schools of the budget disbursement. (6) The District State Treasury disbursed the cash to schools upon receipt of a fund request and the supporting documents. (7) Direct payment in cash to the students (i.e. the households) by schools was based on the number of days they attended school. 23

36 CHAPTER 4 24

37 Main Findings This section presents the main findings of the pilot PETS regarding the shortcomings in the design and implementation of the program. The first part discusses delays in the preparation and execution of the budget, Part 2 identifies potential leakages, Part 3 discusses equity issues and Part 4 presents other inefficiencies in the design and implementation of the program Delays A review of documents and records provided by the Dien Bien provincial DOET and the DDOETs in Muong Cha District and Dien Bien Dong District revealed big differences in the timeline according to the regulations and the actual implementation of D.112 during the last four semesters (see Table 7). One of the reasons for this was the difference between the school year and the fiscal year. The budget process for D.112 in Dien Bien Province followed the school year, which started in September and ended in late May of the following year. 19 In addition, instead of budgeting once a year, the process of issuing budget guidelines, submitting the budget needs, and making a decision on budget allocation at the provincial, district and commune level in Dien Bien Province was done twice a year, once for each semester. Table 7. Time disparity between the official and the actual timeline for D.112 in Dien Bien Province (school years and ) Process Main activities Timeline (as per State Budget Law 2002) Timeline for D.112 in fiscal year 2009 Timeline for D.112 in fiscal year 2010 Timeline for D.112 in fiscal year 2011 Budgeting guidelines 1. The Prime Minister issues the directive on state budget preparation for the coming fiscal year. Before 31 May of the previous fiscal year. 6 June June June The school year starts in early September. The first semester of the academic year is equivalent to the four last months of the fiscal year (from September to December), and the second semester is equivalent to the five first months of the following fiscal year (from January to May). 25

38 Process Main activities Timeline (as per State Budget Law 2002) Timeline for D.112 in fiscal year 2009 Timeline for D.112 in fiscal year 2010 Timeline for D.112 in fiscal year 2011 Budget preparation phase 2. The MOF issues guiding circulars providing detailed guidelines and timetables about the budget preparation process to the ministries (including the CEMA) and provinces. Before 10 June of the previous fiscal year. 20 June June June The PPC instructed the DPC and then the DPC instructed the line agencies (e.g. DOET) and the CPC. The line agencies instructed their subordinates (e.g. schools) to make their budget plans. Once a year. The PPC decided the time frame. The Dien Bien DOET delivered instructions twice each fiscal year, in March/April for the first five months of the fiscal year, i.e. the second semester of the school year, and in August/September for the last four months of the same fiscal year, i.e. the first semester of the next school year. 4. The CPC or its subordinates (e.g. schools) submitted budget estimation to the DPC or district level supervisors, then the DPC consolidated and submitted to the PPC. Once a year The PPC decided the time frame. Schools submitted budget estimation twice each fiscal year, in March/April for the first five months of the fiscal year, i.e. the second semester of the school year, and in August/September for the last four months of the same fiscal year, i.e. the first semester of the next school year. 5. The PPC submitted budget plans to the MOF, MPI and CEMA. Before 20 July of the previous fiscal year. n/a n/a n/a 6. The MOF worked with the central and local agencies, consolidating and building the national budget plan. No later than 31 July of the previous fiscal year. n/a n/a n/a Review, approval and allocation of the state budget by the National Assembly 7. The Government submitted the draft state budget to the National Assembly for review and appraisal. No later than 1 October of the previous fiscal year. n/a n/a n/a 26

39 Process Main activities Timeline (as per State Budget Law 2002) Timeline for D.112 in fiscal year 2009 Timeline for D.112 in fiscal year 2010 Timeline for D.112 in fiscal year The National Assembly convened, discussed, asked questions, and approved the state budget and its allocations. No later than 15 November of the previous fiscal year. 8 Nov For D.112, the MOF allocated to Dien Bien twice, on 1 April and on 16 May Nov For D.112, the MOF allocated to Dien Bien twice, on 21 April and on 8 December Nov For D.112, the MOF allocated to Dien Bien once on 24 March Review, approval and allocation of the state budget at the localities, by the Provincial People s councils 9. The PPC decided the budget apportionment for the sub-national levels. Once a year before 10 December of the previous fiscal year. For D.112, the Dien Bien PPC allocated a budget twice each fiscal year, in April/May for the first five months of the current fiscal year, i.e. the second semester of the school year, and in October/ December for the last four months of the fiscal year, i.e. first semester of the next school year. 10. The District People s Council decided the district budget allocation plans. Once a year before 20 December of the previous fiscal year. The DPC and the DDOETs in Dien Bien decided twice each fiscal year, in June/July of each year for first five months of the fiscal year, i.e. the second semester of the school year, and in December of the current year/ January of the following year for the last four months of the fiscal year, i.e. the first semester of the next school year. Source: Consolidated from legal documents and records from the MOF, the Dien Bien DOET and the DDOETs in Muong Cha and Dien Bien Dong Districts. Among the other main reasons that could explain the time disparities are: (1) a possible attempt to ensure that only children who were attending school were on the recipient list, and (2) a delay in verifying that recipients were from eligible poor households as poverty status was reviewed annually at the beginning of the fiscal year (in the middle of the school year). These revisions of the budget allocation procedures had direct impacts on various budget stages. The discrepancies observed at each stage in the budgeting and allocation process are reviewed below. The budget request preparation and submission process Step 1: The Prime Minister issued the directives regarding the preparation of the state budget for the coming fiscal year Data collected over the last two years showed that the directives were issued as regulated. 27

40 Step 2: Release of budget instructions It was observed for most semesters and in both of the surveyed districts that it took about one week for budget instructions to be sent from the DDOET to recipient schools at the beginning of the process. There was only one exception, in Muong Cha District, when it took almost one month in the second semester of Although there are no specific requirements that address this issue, in the context of the later-than-intended transfer from the central level, it is a step which can be improved in similar situations in the future. The timeline for budget initiation varied largely between semesters. Budgeting for the first semester started at the beginning of the school year, while budget preparations for the second semester started around the middle, or even at the end, of the school year (19 March 2010 and 6 April 2011). This is one of the key reasons for the observed delays in the budget allocation process, particularly for the second semester. Step 3: Delay in budget request preparations at the school level The actual process for budget preparations at the commune (school) level took a longer time and greater effort, mainly due to the complicated procedure for identifying eligible children. As per government regulations, each fiscal year commune authorities assist the DDOLISA in the monitoring of each household s poverty status and providing poor households with a yellow card. In principle, every poor household with a yellow card is eligible for all the benefits provided to the poor by the Government of Viet Nam without additional requirements for verification. However, in Dien Bien Province, in order to be included on the list of those eligible to receive D.112 benefits, households were required to submit three notarized copies of their yellow card to schools. 20 As per government regulations, in most provinces, the offices of the District or Commune People s Committee was the only place that provided notary services. Moreover, not only was it required that the list of students potentially eligible to receive funds be verified by the communes before submission to the DDOET, the DDOET had to cross-check this list against the list of poor households recently updated by the DDOLISA as well. In Muong Cha District, school managers reported that it took them a long time to go to the district center and work along with both DDOET and DDOLISA officers to complete this verification step. It was reported by survey respondents that the budget preparation process was time-consuming and a waste of resources for all stakeholders, particularly poor parents and school managers. More evidence and a detailed discussion of this will be presented below. Steps 4 and 5. Delays in budget preparation at the district and provincial level The process of compilation and budget consideration from the district to the provincial and then to the central level was also long, in general about one month according to documents obtained from the Dien Bien Dong DDOET. 21 Some other problems rendered the process of beneficiary identification more complicated and lengthy, in particular the misspelling of students names; the late provision of the yellow card to school officials; 20 This was clearly specified in the guidelines from the Dien Bien DOET as well as the DDOET, e.g. Instruction 1817/ SGDĐT-KHTC issued by the provincial DOET of 11 Sept. 2010; Instruction 345/PGDĐT-HD of 13 Sept by the Muong Cha DDOET; and Instruction 496/PGDĐT-KTTV issued by the Dien Bien Dong DDOET on 16 Sept In addition, more detailed guidelines from the district were given to schools, e.g. Arrange the yellow cards in the order the people appear on the list of beneficiaries and Only notarized copies of the yellow card made after 1 March 2010 will be accepted for verification (Muong Cha District). 21 From 1 October 1 to 29 December 2010 for the first semester of and from 26 April 2011 to 30 May 2011 for the second semester of

41 and the inability to have the yellow card photocopied and notarized personally (and leaving it to the schools to take care of). Moreover, differences in the budgeting process were observed across districts, indicating a misunderstanding of the budget instructions. For instance, in Dien Bien District, 22 the DDOET asked for a list of poor children from the poor families and then sent it to the schools to be checked against the list of poor children attending school. This process is reported to have saved considerable effort identifying beneficiaries. In Muong Cha and Dien Bien Dong Districts, however, almost all the schools (80 % and 95 % respectively) checked the list twice a year, first with the CPCs and then with the DDOET. Even in the same district, the process was sometimes different. Table 8 presents the frequency that schools checked the beneficiary list with key agencies. Five out of 27 schools in Muong Cha District checked the list against DDOLISA records twice a year, two said they checked only once per year, and the remaining schools did not report having done so. At the same time, in Dien Bien Dong District, 31 of the 37 schools surveyed checked once with the DDOLISA, and one school reported having had to check four times for supplementary funding (see Table 8). Table 8. Frequency that schools verified the list of eligible beneficiaries with the key agencies Checked the list with the CPC Checked the list with the DDOET Checked the list with the DDOLISA Muong Cha District Dien Bien Dong District Muong Cha District Dien Bien Dong District Muong Cha District Dien Bien Dong District Once a year % 7.4% 10.8% 7.4% 2.7% 7.4% 8.1% Twice a year % 59.3% 78.4% 74.1% 54.1% 18.5% 83.8% 3 times a year % 7.4% 0.0% 7.4% 0.0% 0.0% 0.0% 4 times a year % 4% 3% 4% 0% 0% 3% Never % 22% 8% 7% 43% 74% 5% Total Source: Data from the survey of 65 schools in Muong Cha and Dien Bien Dong Districts. 22 Districts visited during the institutional analysis phase. 29

42 The budget allocation process and the problem of delays Steps 8-9: Allocation from the provincial to the district level Examination of financial documents and records shows that it took the province about 20 days to allocate the funds after receiving them from the central government for both the second semester of and the first semester of An extreme case was the second semester of , when the province retained the funds for almost 67 days. In fact this was the last allocation for D.112 as P.135 had ended at the end of 2010 and the central government had advanced the funds before the province had even initiated the budgeting process for school support. Though this was similar to the stipulated time allowed, about three weeks, it should have been done quicker on the fact that all the transfers from the central level were far behind schedule. Steps 9 to 10.1: Allocation from the District People s Committee to the DDOETs The notification process between district agencies added around days to the overall delay. Only for the first semester of was the allocation from the DPCs to the DDOETs in both districts made just two days after notification of the allocation from the province. For the second semester this task took around days. An exception was the Muong Cha DPC, which made the allocation to the DDOET after a month for the second semester of and only after two months for the second semester of This situation requires attention for similar programs in the future as this step is not even mentioned in the general budgeting process. Steps 10.1 to 10.2: Budget allocation from the DDoETs to the schools Compared to the higher levels, the DDOETs in the two surveyed districts were fairly quick in releasing funds to the schools. Two times it was recorded that the funds were disbursed on the same day they were received in Muong Cha District, once the first semester of and again the first semester of ). However, on average it took them up to one week to do so. An exception was again Muong Cha District, where the funds for the last semester of were kept for almost one month (25 days). There was a large deviation in the timeline for budgeting D.112 (see Table 9). According to the State Budget Law, D.112 budget preparation for the next fiscal year should be made prior to July 20 of the current year. Meanwhile, in order to give the authorities in Dien Bien Province enough resources and time to make monthly disbursements to the students the next year, the D.112 budget transfer from the central level should be made to Dien Bien Province prior 31 December of the current year. In practice, however, these procedures were violated, with a delay of three to 12 months at all levels, including the central level. Consequently, the districts and schools in Dien Bien Province could not make the required monthly transfers to students as per the design of Program D.112. Instead, payment was made twice: for the first four months of the school year (the first semester), and then again for the remaining five months (the second semester). 30

43 Table 9. Actual budgeting processes for the four semesters of the and school years in the two surveyed districts 1st semester fiscal year nd semester fiscal year st semester fiscal year nd semester fiscal year 2011 Đien Bien Đong District Muong Cha District Đien Bien Đong District Muong Cha District Đien Bien Đong District Muong Cha District Đien Bien Đong District Muong Cha District Step 2.1: Guidance from the DOET 10 Aug Aug March March Sept Sept April April 2011 Step 2.2: Guidance from the DDOET 17 Aug Aug March March Sept Sept May 2011 Step 3: Schools estimation & the DDOET s consolidation 31 March Oct April 2011 Steps 4 & 5: The DOET s consolidation & submission to the DOF 7 May May 2011 Delay in timeline from step 2.1 to step 5 Delay of about 12 months (should have been made prior to 20 July of fiscal year 2008) Delay of about seven months (should have been made prior to 20 July of fiscal year 2009) Delay of about 11 months (should have been made prior to 20 July of fiscal year 2009) Delay of about eight months (should have been made prior to 20 July of fiscal year 2010) Steps 6-7: Consolidation at the central level n/a Step 8: Advance from the MOF for D.112 n/a n/a 24 April April Dec Dec March March 2011 Step 9: The PPC s allocation notification 22 Oct Oct May May Dec Dec May May 2011 Step 10.1: The DPC s allocation notification 13 Nov Nov May June Dec Dec June July 2011 Step 10.2: The DDOET notified the schools n/a 6 Nov June June Jan Dec June August 2011 Delay in timeline from step 8 to step 10.2 Delay of 10 to 11 months Delay of four to seven months Delay of about 12 months Delay of 3 to 6 months Source: Compiled from records and documents from the DDoETs and all the schools surveyed in Dien Bien. 31

44 The school survey results once again confirmed the long duration of the budget process (see Table 10). When adding up the amount of time it took to go from step three to step 10, the budget approval procedures could take up to 107 days. The only exception was the procedure for the first semester of (44 days) thanks to the availability of a large amount of money carried forward from the previous school year. Table 10: Length of time for budget approval as reported by schools (number of days) according to vouchers/supporting documents and completed forms Muong Cha District Dien Bien Dong District Average Length Number of valid observations * Average Length Number of valid observations * 1st semester n/a n/a 2nd semester st semester nd semester Source: Supporting documents collected from the surveyed schools. Note: * = taken from documents which clearly indicate both budget estimation/ submission dates and notification dates. Muong Cha District Dien Bien Dong District Average Length Number of observations Average Length Number of observations 1st semester nd semester st semester nd semester Source: Authors calculation compiled from forms filled out by schools. Delay in the release of budget resources along administrative levels Funds transferred from the central government to Dien Bien Province Funds allocated by the central government (the MOF) to Dien Bien Province to implement D.112 for the period were delivered in several installments. Although all of the funds (100 %) appear to have been released, this was done in an untimely manner, with each installment arriving late. The delays observed are presented in Table For instance, in , the amount of 32,100 million VND was transferred to Dien Bien Province from the MOF in April 2010, almost at the end of the school year, to cover the needs of all the recipient schools and kindergartens for both semesters of the school year. For the first semester of the school year, 16,312 million VND was transferred in December 2010, again at the end of the semester. 32

45 Table 11. Time D.112 funds were transferred from the central government to Dien Bien Province 1st semester nd semester st\ semester nd semester Funds liberated by the MOF to the province (for all three levels and kindergarten) 32,100 16,342 22,005 Date of receipt 21 April Dec March 2011 Delay Around 5 months 11 months 4 months Source: Provincial DOET reports from Dien Bien, documents from the MOF. The transfer of funds from Dien Bien Province to the districts Disbursement of the support funds from the province to the districts took a much shorter time, about 17 to 66 days, during the surveyed period. When the school year ended in May 2011, the province had to advance money from the local budget (about one third of the total funds needed) to schools while waiting for the budget resources from the central government to arrive. The situation is clearly described in Decision 496/QĐ-UBND of 30 May 2011, when the Dien Bien PPC temporarily mobilized resources from other budget lines while waiting for the disbursement to make the transfers to the poor students. The transfer of funds from the districts to the schools As with the execution of steps 10.1 and 10.2, it was revealed that the DDOETs in the two surveyed districts were fairly quick in releasing funds to the schools. On average it took them up to one week, and twice the funds were transferred on the same day they were received. An exception was Muong Cha District, where the funds for the second semester of the school year were kept for almost one month. A significant delay in the transfer of funds from the schools to the households It was reported in most of the templates collected from the schools that the schools had released the cash to the students as soon as they withdrew it from the treasury. Two thirds of the schools reported having released the funds within 10 days and half of them on the same day. However, a more detailed check against the accompanying documents acquired from the schools revealed some very interesting results: It was almost impossible to release all the funds on the same day, taking into account the long distance from the schools to the treasury offices and the lack of good means of transport by the recipient households. Moreover, a check revealed a large time gap between the dates of cash withdrawal from the treasury and the dates of its disbursement by the schools to the recipients. Of the 35 schools that replied to the template question, 14 of them delayed the transfer for more than 10 days in the two semesters of the school year. In the first semester of the school year, some of them only released the funds after three or four months (see Table 12). 33

46 Table 12. Time gap between the dates of cash withdrawal from the treasury and the dates of its disbursement to the students by the schools (number of days) số ngày 1st semester nd semester Hua Ngai Secondary School 129 Ca To Secondary School 68 Muong Anh Secondary School 42 Ta Te Secondary School 33 Suoi Lu Secondary School 23 Muong Anh Secondary School 13 Ta Te Secondary School 9 N.2 Muong Muon Secondary School 11 Hua Ngai Boarding Secondary School 6 Chieng So Lower Secondary School 11 N-1 Mưong Muon Secondary School 3 Hua Ngai Secondary School 10 Tan Phong Lower Secondary School 2 Tia Dinh Secondary School 5 Muong Luan Secondary School 2 Mưong Luan Secondary School 3 Pa Vat Secondary School 2 Pa Vạt Secondary School 3 Pu Nhi Lower Secondary School 2 Tan Lap Secondary School 3 Xam Man Secondary School 1 Hua Ngai Lower Secondary Boarding School Luan Gioi Lower Secondary School 1 Tan Lap Lower Secondary Boarding School 2 2 Keo Lom Lower Secondary School 1 N-1 Muong Mưon Secondary School 1 Dien Bien Dong Township Lower Secondary School (18) Pu Nhi Lower Secondary School 1 Tan Lap Lower Secondary Boarding School (25) Xam Man Secondary School 1 Huoi Leng Secondary School (120) Nong U Lower Secondary School 1 Pu Hong Secondary School (122) Nong U Lower Secondary School (160) Source: Compiled from the templates provided by schools with 35 valid observations for the first semester and 30 valid observations for the second semester (valid = templates provided both dates of withdrawal and dates of transfer). More importantly, very few schools could provide documents showing accurately documented dates of transfer to the recipient households. Using only the data provided in the templates which were filled out and submitted by the schools management, of the 65 returned templates, there were only 38 date entries for the first semester and 34 entries for the second semester of the school year. In addition, several schools, instead of providing the date of cash transfers to the households, reported the date of the submission of the budget estimation to the DDOET. As a result, computation of the duration (the number of days) between the date of transfer and the withdrawal date sometimes provided negative results (five cases). 34

47 Poor compliance with the general regulations on the timeline for D.112 implementation was clearly observed at almost all the stages and steps of the budgeting process. A very serious time delay was found at the stage of budget preparation. The process for budget needs estimations took much longer than was needed at all the administrative levels. At the provincial level, the shortest recorded time for the budget estimation was around 90 days (13 weeks) and the longest was almost 150 days (five months). At the district level, the time needed for budget preparation and approval was more than 130 days (more than four months) for the second semester of the school year in Muong Cha District (the longest), and around 74 days (about 2.5 months) for the second semester of in Dien Bien Dong District (the shortest). At the frontline level, the process for budget preparation took even longer and required more effort, mainly due to the complicated procedure of beneficiary identification and the discrepancy between the fiscal and the school year. Apart from the delay in the process of budget expenditure preparation and approval, a significant delay was also observed in the funds transfer and cash delivery process of D.112. Evidence from the surveys shows that the most serious time lag was found at the stage of transferring resources from the central government to Dien Bien Province. According to regulations, the eligible funds should have been transferred from the central government to the province before 31 December each fiscal year. In fact, the funds were transferred to Dien Bien Province very late. In fiscal year , the funds for the second semester arrived almost five months later. The delay was much longer (11 months) for the first semester of the school year. For the second semester of the same year, Dien Bien Province had to wait more than four months for the funds to be released (see Table 13). Table 13. Time lag in the budgeting procedure for D.112 Steps Main activities Timeline (as stipulated by the State Budget Law 2002) Lag in the actual budgeting process for D.112 in fiscal years 2009, 2010 and 2011 Step 1 The Prime Minister issued a directive on state budget preparation for the coming fiscal year. Before 31 May of the previous fiscal year. The MOF issued guiding circulars providing detailed guidelines to provinces about the budget preparation process. Before 10 June of the previous fiscal year. Five to 10 days after the stipulated deadline. Step 2 The province (DOET) conveyed instructions to the districts (DDOET) and then the districts (DDOET) provided instructions the communes (schools) regarding the budget request (plan). The PPC decided the timeframe. The process took three to four months for two sets of instructions to be prepared and delivered (one for the fiscal year and one for the school year). Step 3 The schools submitted the budget needs to the district, and then district submitted them to the province. The PPC decided the timeframe. The process took three to four months for two submissions (corresponding to the two schooling semesters). Step 4 The province synthesized the budget plans from the districts to develop the overall provincial budget plan. Before 20 July of the fiscal year. n/a 35

48 Steps Main activities Timeline (as stipulated by the State Budget Law 2002) Lag in the actual budgeting process for D.112 in fiscal years 2009, 2010 and 2011 Step 5 The PPC submitted a provincial budget plan to the Provincial Committee for Ethnic Minority Affairs (PCEMA), the MPI and the MOF. (No date was available from Dien Bien Province.) Step 6 The MPI and the MOF compiled the provincial budget plans into the national SEDP and the budget plan to send to Office of Government and Prime Minister. No later than 31 July of the fiscal year. n/a Step 7 The Prime Minister reviewed and then sendt the national budget plan to the National Assembly for approval at their second annual session No later than 1 October of the fiscal year. n/a Step 8 The province received their appropriated budgets from the central government. No later than 15 November of the fiscal year. Four to five months after the stipulated deadline. Step 9 The PPCc approved the budget apportionment for the district. Before 20 December of the fiscal year. Four to five months after the stipulated deadline. Step 10.1 The DPCc approved the budget apportionment for the DDOETs and the DDOETs made apportionments to the schools. Before 31 December of the fiscal year. Six to seven months after the stipulated deadline. Step 10.2 Funds were transferred from the schools to the households/ students. Every month. Two weeks to six months every semester. Source: Consolidated from existing legal documents, from the records of the DOET and the DDOETs in Muong Cha and Dien Bien Dong Districts and Dien Bien Province. While the process of releasing funds from the provincial level to the districts and from the districts to the schools was found to be relatively timely, a good number of respondents reported that it took a long time for the D.112 support to be delivered by the schools to the households. Some schools reported having released the money to the end recipients after 10 days. Some schools (e.g. Nong U and Hua Ngai) made payments to their students long after the end of the semester, during summer vacation or the following school year. However, results from the household survey portrayed a much more complicated picture- households only recalled having received money without knowing clearly the support is for which year; the time of receipt and amounts of receipt are also quite diversed. Based on the fact that the dates of fund allocation from the DDOET to the schools was made on 10 January 2011 for the first semester in Dien Bien Dong District and on 22 June 2011 and 11 August 2011 for the second semester in Muong Cha District (see Table 9), the following hypothesis can be put forward with the assumption that the two transfers in the school year were made as follows: A full transfer for four months (560,000 VND) for the first semester was made between January and June 2011 for the schools in Dien Bien Dong District and between January and August 2011 for the ones in Muong Cha District. A full transfer for five months (700,000 VND) for the second semester of was made after June 2011 for Dien Bien Dong District and after August 2011 for Muong Cha District. 36

49 The full amount was transferred as there was no mention of any deductions due to students dropping out of school in the DDOETs reports. The results of the calculation are presented in Table 14, which shows that in Muong Cha District the majority of the beneficiaries confirmed receipt of the funds within one month, but it took four or five months for some of them to receive their money. In Dien Bien Dong District, a fifth to a quarter of the beneficiaries recalled that it had taken four to five months for them to receive their money. This delay added to the delay in budgeting and allocation from the central to the local level, which resulted in some beneficiaries receiving their support a year late. Table 14. Time resources were received by respondents in Muong Cha and Dien Bien Dong Districts for each semester of the school year Muong Cha District Dien Bien Dong District 1st semester Month funds were released from district Month funds were received Number of observations Month funds were released from district Month funds were received Number of observations 31 Dec Jan (61%) 10 Jan Jan (27%) Feb (8%) Feb (12%) March (4%) March (13%) April (1%) April (16%) May (8%) May (28%) June (3%) June (4%) July (14%) August (10%) Total 118 (100%) 147 (100%) 2nd semester Month of released from district Month of received Number of observations Month of released from district Month of received Number of observations 22 June 2011 July August (1%) 15 Aug Sept (74%) Sept (40%) Oct (19%) Oct (37%) Nov (2%) Nov (10%) Dec (4%) Dec (12%) Total 212 (100%) 116 (100%) Given that one of the primary objectives of D.112 was to provide timely (monthly) and direct support to poor students to improve their conditions for studying and thereby to improve their school attendance, the delay in the delivery of funds limited the expected outcome and impact of the program. 37

50 4.2. Measurement of potential fund leakage at various levels No official evidence of D.112 fund leakage could be found at this stage. The reason for this was that at the time of the survey the final account settlements for D.112 had not been concluded in either district. However, some serious signs of leakage at various levels of budget implementation were found during this PETS study. Amount of funds transferred from the central government to Dien Bien Province Although funds allocated by the central government (MOF) to Dien Bien Province to implement D.112 for the period were delivered in several installments and arrived late, all of the funds (100%) were reported to have been transferred directly to the lower levels (see Table 15). Table 15. Amount of funds transferred from the central government to Dien Bien Province for D.112 1st semester nd semester st\ semester nd semester Amount in VND 32,100,000 16,342,000 22,005,000 The amount of funds received by Dien Bien Province (for all three levels and kindergarten) Percentage of funds received by the province vs amount liberated by the central government 32,100,000 16,342,000 22,005, % 100% 100% Source: Dien Bien provincial DOET reports and documents from the MOF. Amount of funds transferred from Dien Bien Province to the districts Budget allocations from the central level to the provincial level, from the province to the districts, and from the districts to the schools are reported to have been sufficient to cover the needs of the students (except in 2011). In Muong Cha District, although the exact amounts requested could not be determined through the available documents, the positive balance carried forward every semester is evidence that the funds provided were sufficient. The following balances were consolidated from reports by the DDOET: For the second semester of : 1, million VND For the first semester of : 1, million VND For the second semester of : million VND In Dien Bien Dong District, as seen in Table 16, all the requests for funds for all four semesters were met in full. 38

51 Table 16. Funds released to Dien Bien Dong District over the two school years for all three levels of schools in the district 1st semester nd semester st semester nd semester Budget request (budget needs estimations) The amount of funds transferred by the province to the DPC The amount of funds transferred by the province to the DPC as %age of budget allocation The amount of funds received by the DPC The amount of funds received by the DPC as percentage of funds transferred from the province 3,108,000 4,557,000 2,469,000 3,126,000 3,108,000 4,557,000 2,469,000 3,126, % 100% 100% 100% 3,108,000 4,557,000 2,469,000 3,126, % 100% 100% 100% Source: Compiled from records and documents from the DPC and the DDOETs in Muong Cha and Dien Bien Dong Districts. Budget and funds released from the DPC to the DDOET The total allocation and release for D.112 support in the two districts over the final two school years are summarized in Table 17 for Dien Bien Dong District and in Table 18 for Muong Cha District. While no big time delay was observed in approval of the release of 100% of the total amount allocated to the respective DDOETs, some disparities in the amount released were found. Specifically, the DDOETs released most (92%), but not all, of the resources to the schools in second semester of the school year and the first semester of the school year. In Muong Cha District, excess funds were allocated. A positive balance had been carried forward for D.112 spending at many schools the previous school year ( ), and that was the main reason only 77 % of the funds received in Muong Cha District were released in the first semester of Table 17. Funds transferred in Dien Bien Dong District during both school years 1st semester nd semester st semester nd semester Amount transferred by the province to the DPC Amount received and transferred by the DPC to the DDOET (for all 3 levels) Funds received and transferred by the DPC to the DDOET as percentage of funds transferred by the province to the DPC 3,108,000 4,557,000 2,469,000 3,126,000 3,108,000 4,557,000 2,469,000 3,126, % 100% 100% 100% 39

52 1st semester nd semester st semester nd semester The amount of funds received and transferred by the DDOET to the schools (for all 3 levels) The funds received and transferred by the DDOET to the schools as percentage of funds transferred by the district to the DDOET n/a 4,184,000 2,281,00 3,126,000 n/a 92% 92% 100% Source: Compiled from records and documents of the two DPCs and the DDOETs. Table 18. Funds transferred in Muong Cha District during both school years (in VND) 1st semester nd semester st semester nd semester Amount transferred by the province to the DPC (for all 3 levels) Amount received and transferred by the DPC to the DDOET (for all 3 levels) Funds received and transferred by the DDOET to the schools as percentagage of funds transferred by the district to the DDOET (for all 3 levels) Amount received and transferred by the DPC to the DDOET (for primary and lower secondary) Amount received and transferred by the DDOET to the schools (for primary and lower secondary) Funds received and transferred by the DDOET to the schools as percentage of funds transferred by the district to the DDOET 5,969,000 5,514,000 2,904,000 5,466,000 5,969,000 5,514,000 2,904,000 5,466, % 100% 100% 100% 4,785,000 7,140,000 4,599,000 3,689,000 6,960,000 4,613,000 77% 97% % Source: Compiled from records and documents from the DPC and the DDOETs in Muong Cha and Dien Bien Dong Districts. Release of funds from the district treasury to the schools Results from in-depth interviews showed that the State Treasury system was very efficient and supportive in the fund releasing process. As seen in Table 19, almost all the schools cashed out all of the allocated funds (100% of the schools in Dien Bien Dong District and 99.5% of the schools in Muong Cha District). 40

53 Table 19. Cash withdrawn, spent and carried forward (mil.vnd) 1st semester nd semester st semester nd semester Schools in Dien Bien Dong District The amount of funds liberated by the DDOET to the schools (for primary and lower secondary) Amount withdrawn by schools from the State Treasury Amount withdrawn by the schools from the State Treasury as percentage of liberated amount from the DDOET Amount reported by schools to the DDOET as disbursed Amount disbursed by the schools to students as percentage of amount withdrawn from the State Treasury Difference between the amount disbursed by the schools to the students and the amount cashed out by the schools from the state treasuries (1) The DDOET s official consolidated amount carried forward at the schools (2) The difference between the official consolidated carried forward amount (2) and the actual balance (1) n/a 3,651 1,966 2,755 n/a 3,651 1,966 2, % 100% 100% 2,569 3,229 2,418 3,098 n/a 88.40% % % n/a Schools in Muong Cha District The amount of funds liberated by the DDOET to the schools (for primary and lower secondary) Amount withdrawn by the schools from the State Treasury Amount withdrawn by the schools from the State Treasury as percentage of the liberated amount from the DDOET Amount reported by the schools to the DDOET as spent Amount transferred by the schools to students as percentage of amount withdrawn from the State Treasury 3,689 6,960 4,613 n/a 11, % n/a 13,326 n/a 116% 41

54 (mil.vnd) 1st semester nd semester st semester nd semester Balance between the amount disbursed by the schools to the students and the amount cashed out from the state treasuries (1) The DDOET s official consolidated amount as carried forward at the schools (2) The difference between the official consolidated carried forward amount (2) and the actual balance (1) 1,808 1,078 1, Source: Compiled from records and documents from the DDOETs, the district treasuries and schools. Release of funds from the schools to the households and fund leakage reported by the households As mentioned in section 4.2, regarding the budget process there were two modalities of payment: in cash or in meals. However, 100% of the household respondents in Muong Cha District and 99% in Dien Bien Dong District replied that they had received the D.112 support in cash only. The data in Table 20 shows the form of delivery of the support as perceived by the schools and households. This was confirmed by 100% of the school respondents in both districts who followed the guidelines from the DOET to make cash transfers each semester. Table 20. Forms of delivery of D.112 support as perceived by the households and schools Forms of support delivery Muong Cha District Dien Bien Dong District Total Households Cash transfer In kind (e.g. rice, notebooks) Mixed Schools Cash transfer Meals only Cash transfer and meals Other (specify) Source: Data from the schools and the household survey. 42

55 It is normally easier to track cash support than in-kind support, but given the incomplete of data with poor accuracy and a lack of consistency, it was very difficult for the research team to arrive at a final calculation of the overall leakage of the funds. Instead, making some assumptions based on the results of the household survey and careful examination of related administrative documents and records, the team tried to measure the potential fund leakage, taking into account the data on resources reported by households as having reaching them during the school year. Data from the household survey (a more detailed description of all dates and amounts as received as recalled by the households is presented in Annex 3) revealed that only 66 % of the cases did students report having received cash twice for the school year. About 32% said they had received funds only once. In addition, 10 households (9% in Dien Bien Dong District, 5% of the respondents) reported that their children had received support on a monthly basis. The ones who said that they had received monthly support were mainly from Na Phat Primary School and Na Son Lower Secondary School in Dien Bien Dong District (see Table 21). Table 21. Frequency of cash transfers among the surveyed schools ( ) Once a year Twice a year Monthly Other Number of obs. TH số 1 Mường Mươn 4 (13%) 27 (87%) 0 (0%) 0 (0%) 31 (100%) TH Số 2 Mường Mươn 1 (4%) 24 (96%) 0 (0%) 0 (0%) 25 (100%) THCS PT DTBT Mường Mươn 7 (33%) 14 (67%) 0 (0%) 0 (0%) 21 (100%) TH Số 1 Si Pa Phìn 22 (73%) 8 (27%) 0 (0%) 0 (0%) 30 (100%) TH Tân Phong 10 (38%) 16 (62%) 0 (0%) 0 (0%) 26 (100%) THCS Tân Phong 19 (58%) 13 (39%) 1 (3%) 0 (0%) 33 (100%) TH Huổi Lèng 13 (54%) 11 (46%) 0 (0%) 0 (0%) 24 (100%) THCS PT DTBT Huổi Lèng 5 (38%) 8 (62%) 0 (0%) 0 (0%) 13 (100%) TH Mường Tùng 8 (89%) 1 (11%) 0 (0%) 0 (0%) 9 (100%) TH Nậm He 9 (33%) 18 (67%) 0 (0%) 0 (0%) 27 (100%) THCS Mường Tùng 12 (86%) 2 (14%) 0 (0%) 0 (0%) 14 (100%) THCS Nậm He 5 (28%) 13 (72%) 0 (0%) 0 (0%) 18 (100%) Others schools (not sampled) 14 (61%) 9 (39%) 0 (0%) 0 (0%) 23 (100%) Muong Cha District 129 (44%) 164 (56%) 1 (0%) 0 (0%) 294 (100%) TH Luân Giói 0 (0%) 19 (100%) 0 (0%) 0 (0%) 19 (100%) TH Na Ngua 0 (0%) 16 (100%) 0 (0%) 0 (0%) 16 (100%) THCS Luân Giói 0 (0%) 21 (100%) 0 (0%) 0 (0%) 21 (100%) TH Na Phát 0 (0%) 4 (40%) 6 (60%) 0 (0%) 10 (100%) TH Mường Luân 0 (0%) 13 (100%) 0 (0%) 0 (0%) 13 (100%) THCS Mường Luân 0 (0%) 21 (100%) 0 (0%) 0 (0%) 21 (100%) 43

56 Once a year Twice a year Monthly Other Number of obs. TH Pu Nhi 15 (79%) 4 (21%) 0 (0%) 0 (0%) 19 (100%) TH Nậm Ngám 3 (15%) 17 (85%) 0 (0%) 0 (0%) 20 (100%) THCS Pu Nhi 7 (50%) 7 (50%) 0 (0%) 0 (0%) 14 (100%) TH Sư Lư 4 (29%) 10 (71%) 0 (0%) 0 (0%) 14 (100%) TH Pá Vạt 0 (0%) 21 (100%) 0 (0%) 0 (0%) 21 (100%) THCS Na Son 2 (18%) 6 (55%) 3 (27%) 0 (0%) 11 (100%) Others schools (not sampled) 1 (9%) 9 (82%) 1 (9%) 0 (0%) 11 (100%) Dien Bien Dong District 32 (15%) 168 (80%) 10 (5%) 0 (0%) 210 (100%) All Samples 161 (32%) 332 (66%) 11 (2%) 0 (0%) 504 (100%) Source: Data from the household survey. Continuing with the assumptions and hypothesis put forward in the section above, an estimation was made for the average received amount using all cases of receipt of first or second time and put them in their corresponding months against the dates the funds were released from the DDOETs and put into the schools accounts. The average amount of resources received was calculated based on the average of the entries with reported amounts. Table 22 provides information about the time and the average amount of the transfers reported by 146 beneficiaries who had received funds once and 314 beneficiaries who had received funds twice (a total of 774 transfers) each semester. The results are as follows: Regarding the transfer for the first semester, 61% of the beneficiaries in Muong Cha District claimed to have received cash immediately in January, and the remaining transfers were recalled as having taken place some months later. In Dien Bien Dong District, the beneficiaries were notified only 10 days after the people in Muong Cha District, yet only 40% of the transfers were recalled to take place within one month of that date. More noticeably, the average amount of resources the beneficiaries claimed to have received in Muong Cha District was about 511,000 VND, while in Dien Bien Dong District it was only 385,000 VND. The beneficiaries in Muong Cha and Dien Bien Dong Districts received 91% and 69% respectively of the full amount for the four months of the first semester (560,000 VND). For the second semester, the transfers in Muong Cha District were quite fast, with 74% of the transfers occurring right at the beginning of the school year (September), less than a month after notification from the DDOET (22 August 2011). In Dien Bien Dong District, by September 2011, 40% of the transfers were received, in October 37%, and in November 10%. However, it should be emphasized that notification of the release of funds from the Dien Bien Dong DDOET to the schools occurred on 22 June The resources remained undelivered for three months in 2011 in Dien Bien Dong schools and almost no transfers were noted. The average amount received in Muong Cha District was 618,000 VND, accounting for 88% of the amount beneficiaries were supposed to have received (700,000 VND) for the five months. This was significantly higher than the 543,000 VND beneficiaries in Dien Bien Dong District received, which accounted for only 78% of the funds they were supposed to have received. 44

57 Table 22. Time and average amount of resources received by respondents in Muong Cha and Dien Bien Dong Districts for each semester of the school year Muong Cha District Dien Bien Dong District 1st semester Month released from district Month received Number of observations Month released from district Month received Number of observations 31 Dec Jan (61%) 10 Jan Jan (27%) Feb (8%) Feb (12%) March (4%) March (13%) April (1%) April (16%) May (8%) May (28%) June (3%) June (4%) July (14%) August (10%) Total 118 (100%) Total 147 (100%) Number of observations 93(*) Number of observations 71 Average amount received 510,968 Average amount received 385,141 91% 69% 2nd semester Month released from district Month received Number of observations Month released from district Month received Number of observations 22 June 2011 July August (1%) 15 August 2011 Sept (74%) Sept (40%) Oct (19%) Oct (37%) Nov (2%) Nov (10%) Dec (4%) Dec (12%) Total 212 (100%) Total 116 (100%) Number of observations 213(**) Number of observations 115 (**) Average amount received Average amount received % 78% (*) = only calculated observations with the amount of 0 to 560,000 VND (**) = only calculated observations with the amount of 0 to 700,000 VND 24 Source: Data from the household survey. 24 Hypothesis for estimation: Full transfer for the four months (560,000 VND) of the first semester of should have occurred from January to June 2011 for schools in Dien Bien Dong District and from January to August 2011 for schools in Muong Cha District. Full transfer for the five months (700,000 VND) of the second semester of should have taken place from June 2011 onward for Dien Bien Dong District and from August 2011 onward for Muong Cha District. Transfers should have been for the full amount as there was no mention of any deductions due to students dropping out of school in the DDOETs reports. 45

58 With such results, it is appropriate to conclude that by the time of this households survey on average the fund leakage rate in Muong Cha District in the first semester of the school year was 9% and the rate in Dien Bien Dong District was much higher: 31%. Similarly, the fund leakage rate for the second semester of the school year in Muong Cha District was 12% and in Dien Bien Dong District it was 22%. However, this conclusion should be read with serious caution as the numbers reported by the responding households were mostly recalled and not taken from written records. In addition, many of the respondents were uncertain about the precise dates and amounts of the fund transfers. The poor practice of schools of not returning unused D.112 funds The first sign of fund leakage was observed in the practice of the schools not returning unused funds to the treasury. This could have been partially due to the practice of letting the schools retain and carry forward any positive balance to the following semester. According to officially reported data by the DDOETs, the carried forward amount was quite significant at certain times. For example, in fiscal year 2010, the amount carried forward each semester was more than one billion VND in Muong Cha District and about 700 to 800 million VND in Dien Bien Dong District. According to regulations, the unused funds had to be returned to the treasury right after all the cash transfers had been completed. However, this did not happen. As a result, a considerable amount of unused funds remained out of government hands for several months. Table 23 presents the amounts retained by schools in Muong Cha and Dien Bien Dong Districts. These funds should have been returned to the district treasuries. Table 23. Resource surpluses retained by the schools 1st semester nd semester st semester nd semester Đien Bien Đong District Muong Cha District 780 2, , ,910 1,078,420 1,438,520 70,770 Source: Compiled from financial records and documents from the DDOETs, the district treasuries and the schools in the two districts. In addition, some schools withdrew cash during the summer holidays (see Table 24). The real purpose for withdrawing the cash is unclear. This money was reported to not have been disbursed to beneficiaries in a timely manner. 46

59 Table 24. Number of schools withdrawing the support cash during the summer holiday Muong Cha District Dien Bien Dong District 2nd semester Number of observations (withdrawal dates) Number of schools that withdrew funds before 1 Sept (20%) 10 (60%) 2nd semester Number of observations (withdrawal dates) Number of schools that withdrew funds before 1 Sept (0%) 8 (40%) Source: PETS Dien Bien 2011, from documents obtained at the surveyed schools. Fund leakage in the form of paying service fee to intermediaries More seriously, data from the household survey show other evidence of the risk of fund leakage such as the practice of paying some service fee. In the Dien Bien Dong District, about 70% of the beneficiaries reported having paid an informal commission to intermediaries in 2011 in order to collect their cash support. In Muong Cha District, the proportion of beneficiaries who reported having paid some service fee was much less (30%). However, all the respondents refused to state the precise amount of the fee as well as the name and address of person(s) they made the payment to. Figure 5. Percentage of households that paid a commission for receipt of D.112 support % st transfer 2st transfer 20 0 Muong Cha Dien Bien Dong Source: Data from the household survey 47

60 Fund leakage in the form of beneficiaries paying back some expenses to the schools Furthermore, about 30% of the respondents in Muong Cha District and 80 % in the Dien Bien Dong District said that they had had to pay a certain amount of money back to the school for school expenses right after receiving the D.112 transfer (see Figure 6). The average amount they had to pay was 117, ,000 VND upon receipt of the first disbursement, and this amount was the same for both districts. For the second disbursement, the average amount was around 77,000 VND in Muong Cha District and 55,000 VND in Dien Bien Dong District. Added up, the amount paid back to schools ranged from 170,000 to 190,000 VND. This was equal to one month of support for a student. Figure 6. Percentage of households that reported having had to make a payment back to school for school expenses 90% 80% 70% 60% 50% 40% 1st transfer 2st transfer 30% 20% 10% 0% Muong Cha Dien Bien Dong Source: Data from the household survey. In addition, most of the schools confirmed that they had deducted some money from the support amount for the school s needs (e.g. for school maintenance and repairs or the Parents Association fund). Then the amount schools claimed to have deducted was much less than that reported by targeted households. The maximum amount recalled by schools was 115,000 VND, while the average amount paid by each student was around 40,000 VND per year, significantly less than the amount recalled by households. Since the ultimate objective of D.112 was to provide direct support to improve the standard of living of poor students, deductions taken from the support fund for other purposes can be considered a form of fund leakage. Violation of general accounting procedures - a risk of fund leakage Another possible leakage can be seen in the practice of one of the schools in Muong Cha District, where the process of payment against vouchers was applied. Specifically, the accountant from the school in Muong Cha District asked the students to sign a receipt before receiving the cash. This receipt with signatures was later submitted to the district state treasury to get the cash disbursement, and no funding advance procedures were undertaken. After receiving the money from the district treasury, the accountant had the recipients sign another list as evidence of actual payment. Clearly, this case of the accountant asking school children in Muong Cha District to sign a receipt long before receiving their cash support is cause for concern as a number potential risks of fund leakage could have resulted from the practice. Firstly, the students may never have received money at all as 48

61 their signatures were already on the recipient list formally submitted to the treasury as evidence of having received the cash. With the receipt list already signed by recipients, the accountant could have delayed disbursement to the recipient households or even kept the money for herself. Secondly, this practice could make it difficult for the schools and districts to revise and update the list of beneficiaries. A poorly updated list of eligible beneficiaries can provide a very good opportunity to someone to include either non-poor students or keep the students who have dropped out of school on the target recipient list. 4.3 Equity issues Issue of the coverage of D.112 Although both the Dien Bien provincial DOET and the DEMA reported that 100 % of the children from the targeted poor households were covered by D.112, many households and school managers had concerns regarding the coverage. The survey results indicated that the three most common problems with the program s coverage were (1) a failure to include all the eligible students, (2) the inclusion of ineligible students, and (3) a failure to reach students from households that did not have a yellow card but lived under very difficult conditions. More than 85 % of the school managers in Muong Cha District and about 68 % in Dien Bien Dong District expressed concern about the program s neglect of poor students who did not have yellow cards to show to schools. In Dien Bien Dong District, the issue of including ineligible recipients was perceived by the majority of households and school managers to be a problem (see Table 25). Table 25. Perceived failure in the extension of D.112 coverage Failure in the benefit coverage School Managers perception Muong Cha District Dien Bien Dong District Lack of support for eligible students 40.74% 43.24% Failure to reach students from families that did not have a poor card but lived under difficult conditions 85.19% 67.57% Failure to restrict non-poor students from receiving D.112 support 55.56% 64.86% Source: Data from the household and school surveys. A total of 115 respondents, 76 in Muong Cha District and 39 in Dien Bien Dong District, claimed not to have received support although they all had a yellow card in 2011 and regularly attended school. The percentage of eligible students who did not receive support was approximately 20%, and in the case of Huoi Leng School it was 38% (see Figure 7). Looking only at the 24 schools in the eight communes sampled for the household survey, the households reported that 82 students had not received funds (about 20%). Among those, 33 students were included as beneficiaries of the program on the official list of recipients used for sampling. For the remaining 49 students, the reasons for not receiving the funds were either because they had not been identified as eligible beneficiaries or they had not actually attended school full-time. 49

62 Figure 7. Percentage of poor students who did not receive any support from D.112 although they regularly attended school, as reported by households All sample Muong Cha Dien Bien Dong THCS Pu Nhi THCS Na Son THCS Luan Gioi THCS Mường Luân TH Pá Vạt TH Pu Nhi TH Mường Luân TH Sư lư TH Nậm Ngám TH Na Phát TH Na Ngua TH Luân Giói THCS Tân Phong THCS Nậm He THCS Mường Tùng THCS PT DTBT Huổi Lèng THCS PT DTBT Mường Mươn TH Nâm He TH Mường Tùng 12% 17% 20% TH Huổi Lèng TH Tân Phong 38% TH Số 1 Si Pa Phìn TH Số 2 Mường Mươn... TH Số 1 Mường Mươn 0% 10% 20% 30% 40% 50% Source: Data from the household survey. In contrast with the issue of eligible beneficiaries not receiving D.112 funds, the household survey indicated that 16 of the 504 recipients (3%), eight beneficiaries in each surveyed district, reported having received support although they were no longer attending school. 10 of these students had left school during the previous 12 months and six had been out of school for more than a year. A more detailed investigation revealed that of those 10 cases, only four had received support money for the second semester of and the first semester of Two others had received money in October 2011 (700,000 VND for five months, very likely for the second semester of ). Although this number is not very significant (3%), the figures indicated still certain drop-out occurrence despite the support. Besides, they induce the need for closer supervision of pupils actual attendance at schools as the base for determining the level of support so as to contribute to the fulfillment of D112 s objective. 50

63 Issues of imprecise identification of beneficiaries The issue of precision in identifying potential recipients of support from the program was found to be critical. Heavy reliance on the yellow card in the process of identification and selection of beneficiaries led to the risk of leaving out eligible students as well as some over-reporting of the number of potential recipients. Apart from neglecting the eligible students due to a failure to reach them, it was reported during team field visits to both Dien Bien Dong and Muong Cha Districts that there were some households that were no longer poor but had kept their yellow card and continued to receive support from the program. Consequently, in some schools and their communes, the list of candidates for D.112 support included people who were not poor. Then another problem was the inclusion of students who had dropped out of school. Among the sample, 16 had stopped going to school, 10 of them had been out of school within 12 months and 6 had stopped more than 12 months ago but continued to receive support from the program. 4.4 Other shortcomings Overly-complicated beneficiary selection procedures The procedures for identification, selection and approval of eligible children turned out to be rather complicated and overlapping. This was widely noticed by both households and schools during the survey. The widely implemented requirement to have the yellow card photocopied and notarized at the commune and school level made it difficult for poor households, especially those from remote areas, to prove their eligibility for the benefits offered by the program. Such complicated procedures resulted in a waste of time and resources for teachers and commune authorities as well. More than 80% of the households in Dien Bien Dong District and almost 40% of the households in Muong Cha District complained about the complexity of the procedure and viewed this as a very serious problem. Figure 8. Perception of households and school managers on the seriousness of the issue of complicated procedures 100% 80% 60% 40% 20% 0% Hhds in Muong cha Hhds in Dien Bien Dong all school managers Very little a little bit evegare Large Very Large Source: Household and school surveys. 51

64 Box 5.1: Difficulty caused by requirement to have the poor cards photocopied and notarized Nam He Primary School is about 20 km from the commune center, but it takes more than an hour by motorbike to reach it. From there, the road is inaccessible by car. When there is unfavorable weather, walking is the only option. In recent years, in order to prepare the profiles of students eligible to be put on the list for benefits provided by D.112, the school manager had to collect all the yellow cards of the students and carry them in a heavy package to the commune PC office to be notarized, sometimes at personal expense. The school manager is still in debt to the commune authorities for the expense of photocopying and the notary fees. Xa Tong Commune is a Hmong commune located 40 km from Muong Lay Town and it is 1,500 m above sea level. Every semester when the time for preparation of the eligibility list comes, the commune PC office is often filled with poor Hmong parents from all four schools (one kindergarten, two primary and one lower secondary school) who need to have their yellow cards notorized. The only available photocopy machine in the PC office frequently breaks down, causing many parents to rent a motorbike to go to Muong Lay Town to make copies. This process normally costs around 150,000VND and takes an entire day. Source: From in-depth interviews with managers at Nam He School and Sa Tong 1 Primary School in December In addition, the requirement to check and store the photocopies of the yellow cards of poor households by district authorities (e.g. DDOLISA) was reported to be not only time-consuming but also no longer meaningful as the government officers at the district level often did not know the precise number of members in a household or their age or names. Instead, the commune level authorities should be responsible for the credibility of the beneficiary list. Use of the support funds at the household level The direct cash transfer modality was reported as being good in providing the beneficiary households with an opportunity to make decisions. However, this does not mean that there were no cases in which the transferred cash was badly used by the parents to satisfy other needs rather than the needs of the targeted children as specified in the program regulations. Structured conversations between the study team and representatives from the communes (e.g. Muong Muon Commune in Muong Cha District) and some local people in Dien Bien Dong District revealed that there were a number of cases when parents took the cash and spent it on general household needs and even cultural needs. The survey results show that more than 56% of the respondents in Dien Bien Dong District answered that they had spent some of the target cash on the general needs of their families. The number of households misusing the cash support for general needs was lower in Muong Cha District, but still very high (around 20%). It was revealed from the household survey that the main general household expenditures included clothes, food, learning aids, and items for healthcare and education (see Figure 9). A total of 94 % of the households in both districts confirmed that they had spent most of the money on clothing, followed by supplies for education (78% in Muong Cha District and 85 % in Dien Bien Dong District) and food (30-35% in both districts). However, some of them admitted that some money was spent on cultural needs such as buying a gift for a neighbor s wedding or on alcohol. 52

65 Figure 9. Spending pattern of D.112 funds Spending items in Muong Cha district Spending items in Dien Bien Dong district 4, 1% 1% 158, 44% 2, 1% 131, 37% 43% 1% 39% 50, 14% 12, 3% 16% 0% Education Health care Food Clothes Culture Others Source: Data from the household survey. User satisfaction and issue of limited access to information about the program Despite significant identified shortcomings including delays in cash disbursements, lower received amount compared to the official amount stipulated by the Decision 112, complex procedures for identifying, selecting and approving eligible children, it can be seen in Figure 10 that most households are still satisfied with the program- 77% are satisfied and very satisfied. Dien Bien Dong citizens have higher satisfaction ratio than those of Muong Cha- 85% and 70%. Figure 10. Households satisfaction with D % 80% 60% 40% 20% 0% Muong Cha Dien Bien Dong All sample Very satisfied Satisfied So so Not satisfied Completely dissatisfied Source: PETS Dien Bien 2011, household survey Why and how could this happen? One very applaudable reason was the fact that very little information about the program was provided to the public in general and to the target group in particular. Only around 10 % of the respondents in Muong Cha District and 30 % in Dien Bien Dong District reported having been informed about D.112 before receipt of the support money (see Figure 11). 53

66 Figure 11. Percentage of recipients that had been informed about D.112 before each transfer % 20 Before the first transfer Before the second transfer Muong Cha Dien Bien Dong Source: From the household survey. Similarly, only 60 pecent of the surveyed households knew about D.112. About 44% of the households in Dien Bien Dong District knew something about D.112, and 72 % of the respondents in Muong Cha District had heard of the program (see Figure 12). Figure 12: Knowledge of respondents about D % Muong Cha Dien Bien Dong All sample Source: Household survey These data and figures indicate that the majority of the households were not given clear information about the objectives, targets, and operational and financial procedures of D.112. Even though it was very clear that potential beneficiaries had to have a yellow (card, up to 10 % of the households in Muong Cha District and 19 % of the households in Dien Bien Dong District thought that this criterion applied to all poor households (see Figure 13). 54

67 Figure 13. Reasons households were eligible for D Percentage We have poor card I am very poor school selected us missing 0 Muong Cha Dien Bien Dong Source: Household survey Such relatively low level of knowledge about the program, particularly among the households in Dien Bien Dong District, explains the contrast in lower level of receipt but higher level of satisfaction, then the discrepancy in spending patterns in Dien Bien Dong versus Muong Cha as described in the previous section. Inconsistent reporting of the received amount and the time of its receipt is another indicator of a lack of sufficient knowledge by beneficiaries and their families of the objectives and targets of the program. This issue brings lots of implications for information scheme so as to enhance self-monitoring of other similar cash transfer programs Issue of supervision and accountability Supervision and inspection It was clearly stated in D.112 that the PPC in provinces targeted by Program 135-II were responsible for, among other things, regular monitoring and supervision to ensure that appropriate policies were applied rightly to appropriate households, without any losses or corruption. In Dien Bien Province, although the monitoring was reported to have been made through a financial revenues/expenditures scheme pursuant to regulations and guidelines from the DOF, no independent or internal auditing was organized for the program. Moreover, as experienced by the study team, not many annual operational, statistical, accounting and other regular or irregular reports could be found for the purpose of monitoring at the provincial level. Some supervision activities related to D.112 were carried out by provincial authorities, often in the form of field visits or a public hearing, but in an irregular manner. The DEMA reported that as far as they knew, two supervision missions had been conducted, and they had covered other projects and programs at the same time. At the district level it was reported that supervision and inspection were carried out very often by the DOET and the DOF, and even by the local inspectorate. However, very few official reports have been made available to public. When asked by the study team, the DOF and the DOET in both districts could not provide any documented evidence related to the activities. Regarding supervision at the commune level, only one household surveyed in Muong Cha District reported having received an inspection visit. The other 292 households replied that they had never before heard about the inspection missions or received a visit from inspectors. 55

68 Data management As per the regulations, administrative data, including data on the number of eligible students, financial support transfers and cash disbursements, should be properly maintained by frontline authorities and regularly (annually and quarterly) reported to the provincial level. It was found, however, that formats for data management were very different from one school to another, and in Dien Bien Dong and Muong Cha Districts. Thus, it was almost impossible to make a direct consolidation or comparison of the data sets obtained from these entities. Another challenge faced by this tracking survey was the availability and accuracy of the relevant financial records. As discussed above, many schools could not report the precise dates of transfer to students simply because the list of recipients did not bear any specific dates. The quality of the records of cash distributions was low and the data was often incomplete. Financial data management seemed to vary from department to department in the same districts and between districts. The lack of precisely documented times and the amount of the support received made it very difficult for the program to be accountable. The problem became more severe at the frontline and household level. None of the surveyed households could prove how much money they had received and when. Thus, they could neither track the arrival of the support funds for each of the school years nor calculate the total amount of the support their children were eligible to receive. By recalling (rather than based on proper records), households very often mistook the D.112 assistance with resources from other support programs. Checking through supporting documents at the schools helped confirm some poor practices: Many schools failed to precisely report the date of cash transfers to the end recipients as they had not kept clear records (Table 26). The list of benefit recipients provided only their names and signatures without any dates. Therefore, proof of the precise dates of transfer of the money to the end users could not be found, not even at the schools which had kept regular records of the cash transfers. 25 Table 26: Number of schools with documented dates of cash disbursement to students Muong Cha District Dien Bien Dong District Number of schools that provided documents Number of valid observations with vouchers showing both dates schools withdrew cash and dates they transferred it to students Number of schools that provided documents Number of valid observations with vouchers showing both dates schools withdrew cash and dates they transferred it to students 1st semester nd semester st semester nd semester Source: Supporting documents provided by the surveyed schools. 25 As discussed in the above section, some schools, instead of providing the dates of the cash transfers to households, reported the dates of the submission of the budget estimation to the DDOET. This was the situation in five cases, negatively affecting the calculation of the difference between the transfer dates and the withdrawal dates (see Table 12). 56

69 Lack of settlement of the final accounts of the program In terms of overall management, by the time of this study (December 2011), six months after implementation of D.112 had finished, only one final account settlement for the spending in 2009 had been completed. According to the Law on State Budget, the official amount of the budget allocation for the following budget period can be publicized only after the final account settlement report has been submitted and approved by the People s Committee or after the spending agencies have submitted dossiers and vouchers to the state treasuries. This should be done no later than 25 January of the following year. These tasks had not yet been carried out in Dien Bien Province at all levels for budget allocation for D.112, neither for 2010 nor for the first five months of The amount thus far released was considered to be a fund advance only. Therefore, it is difficult to make any final conclusion on the final use of the funds, and, understandably, no conclusion can be firmly made concerning possible leakage of support funds as the accounts are not yet closed. Poor collaboration between the main government stakeholders There was very little evidence of collaboration between different local government departments. At the national and accordingly, at provincial level, D.112 was designed to be implemented by a number of issue-related agencies such as the DEMA, the DOET, the DOF, the DPI and the DOLISA. However, in Dien Bien Province, cooperative relationship mechanisms between the Program Management Board and representatives from the provincial departments was not well implemented and articulated. Due to the shortage of capacity which includes both issues of quantity and quality as mentioned in section 3.3 Organizational arrangements for D.112, provincial DEMA in particular could only modestly involve in issuing guidelines and a budget allocation plan, and participated very little at the budget execution stage. At the district level the situation was the same. While the program was actively implemented by the DDOET, other service-related departments such as the DDEMA and the DDOLISA in Dien Bien Dong District replied that they had not been well informed about the program. Moreover, although assigned to be implementing agencies of D.112, in reality, the DDEMAs at the district level played only a very small role in the budget expenditure planning process. A representative of the Dien Bien Dong DDEMA said that the only thing he was interested in was getting some data on the support of ethnic minority children for annual report writing, and in Muong Cha District the director of the DDEMA had little to say about D.112. Similarly, despite the mandate to consolidate financial data and prepare final accounts for submission and report to the People s Committee, the actual role of the District Division of Finance and Planning (DDPI) was quite limited. Evidence of functional inefficiency was the severe delay in the final settlement of accounts, which had not been completed in either of the surveyed districts at the time this PETS was conducted. This in turn caused significant difficulties when monitoring the program and conducting a performance evaluation. In a nutshell, it can be seen clearly that the lack of capacity prevented DEMA from accomplishing monitoring task to the implementation of P135 in general and D112 in particular as required in Decision 588/UBND-NN by the PPC, 16 June Together with the limited functional efficiency of related departments, these led to the inability of PPC to fulfill the prime task of overall management and supervision defined by Decision 112 itself. Dien Bien PPC has very few reports and little information on the quality of implementation of D112, not to mention related issues of malpractices. 57

70 CHAPTER 5 58

71 Concluding remarks and recommendations 5.1. Concluding remarks D.112 was an important component of P.135, which was designed to support poor and disadvantaged boarding school children who were living in the extreme difficulties communes in mountainous and ethnic minority areas. The designated ultimate purpose of D.112 was to improve the enrollment and attendance of those school children, especially those in the most remote areas. While the type and modality of the direct cash transfer was generally acknowledged as being a good instrument in providing the beneficiaries with an opportunity to make choices, some issues related to financial aspects of its execution were found. As analyzed and revealed in Part 4 above, the key issues of implementation of D112 in Dien Bien include the followings: poor compliance with budgeting regulations at the stage of budget preparation causing significant delay from the very beginning; certain time delay in the process of budget allocation along administrative levels from province to districts, then schools, which made it a failure to deliver the resources to the program s end recipients in a timely manner as regulated; high risk of fund leakage in the process of executing the budget expenditure with certain incorrect accounting processes, the charging of commission and collecting of school expenses upon the payment of the cash transfer; Very limited knowledge of the beneficiaries and their families on the objectives and the regulation of the support. As a consequent, self-monitoring of the program implementation has been absent, leading to even quite high satisfaction despite of all delays and possible leakages of the support delivery; In addition, inadequate equity reflected in the failure to reach eligible students, ineffective use of the support fund, and a lack of an effective mechanism for ensuring accountability and transparency were other inefficiencies that limited the effectiveness of the program. 59

72 For effective solutions to these limitations, the Government should make continuous and strong efforts at the local level to attain more comprehensive improvements in all related matters. Since many of these changes cannot occur at the same time, the goal for improvement should be process-oriented, incremental and harmoniously coordinated with public budget reforms at the national level Recommendations D.112 as a social assistance program was formally completed in However, a number of similar policies and programs are currently being implemented by the Government such as Decree 49/2010, which is similar to D.112 to a certain degree as it targets all the students in P.135 communes. Those interventions could be improved if attention were paid to the lessons leant from D.112. (1) Improve the existing public budget spending system at the local level D.112 gave great discretion to local authorities to identify the best ways to implement the public budget expenditure. Thus, it is highly recommended that authorities in Dien Bien Province take appropriate measures to deepen understanding of the inefficiencies and limitations in the execution of D.112 and to establish a transparent public budget administration system for a more effective allocation of the budgetary resources to the lower administrative levels. When doing this, such important indicators as precise identification of beneficiaries, timeliness and volume of support delivery, and improvement in achievement of the program outcomes should be taken into account. High accuracy in the identification of beneficiaries, adequate equity in program coverage, well-defined types and modalities of assistance and high appropriateness of the public assistance, high quality of the transfer should be strictly ensured. The procedure for identification, selection and approval of eligible children should be simplified such as removal of the requirement to have the yellow card photocopied and notarized as this had made it costly and complicated for certain households to prove their eligibility to the public services. For Decree 49, this implies removal of the requirement to have the residential book photocopied and notarized. Checking and verification by commune authorities might be sufficient provided parallel measures are implemented to ensure accountability. (2) Take decisive action to prevent of all of the fund leakage and loss In order to ensure that the program s limited resources reach the targeted beneficiaries without any leakage, a number of decisive actions should be undertaken, including: Take strong measures to immediately stop the practice of making beneficiaries sign receipts before being paid. Halt the practice of paying a commission to intermediaries in order to collect the cash transfer support. Exercise more control over the practice of deducting part of the cash support to cover school expenses. Strengthen compliance with government regulations concerning the return of inactive funds by schools as this is potentially a serious source of financial loss. Appropriate sanctions and penalties should be applied to make sure that all withdrawn money is disbursed to the beneficiaries by the frontline suppliers in a timely manner. Raise awareness and increase public participation to enhance the knowledge of households 60

73 of the objectives, rules and operational procedures of the program in order to reduce the risk of leakage of the limited funds. Possible channels could be parents associations, commune officers and village heads, who would be capable of overseeing the whole process. Final accounts settlement for the spending each semester should be carried out at all levels pursuant to the State Budget Law. Completion of settlement of final accounts is strongly needed to clarify the results and thus enhance accountability and transparency during implementation of the program. A strong and reliable system of monitoring, auditing, and reporting of the program s activities is a crucial determinant of its credibility and hence its sustainability. Internal control should therefore be exercised by means of legal, financial, and performance audits. The purpose of internal control is to ensure that the program achieves its intended results (performance auditing), and that it is protected from risks such as misuse, waste of financial and other resources, fraud and error, unsatisfactory accounting records, and the failure to execute decisions in an effective manner (legal and financial auditing). It is also recommended that program regulations clearly indicate the duty of the local authorities and the schools to report publicly and widely the nature of the program s activities by means of quarterly and annual reports. Furthermore, for each transfer of resources, the related authorities and schools should report at the very least the name of the beneficiary and the amount and date the financial support was transferred and received. Reports by independent auditors should also be made available upon request. (3) Ensure transparency during implementation of the program Increased access to information should be achieved to increase transparency during implementation of the program. Providing better information to the households, the end users of the benefits, could help improve the program outcomes. People should know that the funds are not simply given to them, but that they are meant to support the improvement of diet, other living conditions and school materials for their children, and that this money should not be used for general needs or other purposes such as gifts and alcohol. During the course of implementation, service facilities, particularly local authorities, should adopt further measures to raise the awareness of the beneficiaries in order to increase transparency and effective use of the support. A large number of poor households had a TV (50%) or a cell phone (45%), so the media (e.g. local TV) and directly through local authorities would be feasible and effective channels for transmitting information. (4) Enhance public participation Enhanced public participation could be achieved through formalizing participation and consultation activities in the budgeting process, particularly the transfer steps. Taking into account the low level of knowledge of the poor and minority households, households should keep clear records and receipts of fund transfers, and during transfers at schools or any other service delivery facility the poor should be represented by local authorities to ensure transparency and effective supervision of the use of such support. 61

74 (5) Improve supervision and accountability Making the implementation of the program transparent should be addressed from the very first stage of the program, the stage of program implementation planning, throughout the stages of its execution and at its conclusion. In addition, for the transparency of the program as a whole, progress at each stage should be reported with specific performance indicators. Budget allocation and budget allocation norms should be published via appropriate channels at all levels, especially at the frontline level for service providers and service users. The filing system and records at all management levels should be improved. This is particularly important at the household level. Due to the lack of a formal banking transfer system, records that show the dates of transfers and how much was received should be kept by both the schools and the targeted households as evidence of transactions. Such tools can also raise the awareness of the households on the objectives and targeted beneficiary of the program. It is strongly recommended that supervision, monitoring and evaluation of the program and its social and poverty impacts be reinforced in accordance with existing regulations. The DEMA and the DDEMA should be assigned as appropriate agencies to monitor the program in terms of budget allocation, disbursement mechanisms, and payment to beneficiaries. School boards and citizen inspection committees at the commune level could be mobilized to monitor payment and utilization of the provided resources. An organization independent from the program s executive management, the program-related government agencies and the project implementers should be hired to carry out an evaluation and prepare a relevant assessment report. (6) Improve the management structure and coordination Among the various issues faced by D.112, the management structure, capacity and coordination deserves serious attention. The credibility of programs like D.112, as well as its long-term sustainability, crucially depends on these things. The capacity of the local government staff at all levels should be strengthened, possibly by further training and retraining, to ensure that they have proper professional qualifications, good command of legal knowledge, and that they respect and comply with the law. Training content should be realistic and pertain to local characteristics, and it should not be a lengthy process. During the training process, in addition to the improvement of specialized knowledge and qualifications, the right attitude and a sense of responsibility for the work and the public should be reinforced. For any future similar cash transfer programs, focal units such as DEMA and DDEMA and should be the first targets for capacity building as these agencies have complex responsibilities and work with a large number of service providers and target users. Given their important role in connecting the state support to the service providers (schools) and end users (households), it is highly recommended to pay more attention to strengthening their capacity to fulfill their mandates, including financial management skills and incentive structures. In addition, there are two important determinants of the program s success insofar as the management structure is concerned: Management of the program should ensure a wide and carefully balanced representation of the key stakeholders. Thus far the voice of households, the end recipients of the fund support, have not been heard as they have not had a chance and space to voice their opinion. 62

75 Results of the survey revealed the very low capacity of several positions (accountants and even some directors of local authorities departments/divisions), especially regarding aspects of financial management. It is therefore highly recommended that all individuals associated with the management of the program s funds be held accountable to the highest standards of good professionalism and good moral behavior. (7) Suggested themes for further study Due to the limit of time and resources, there are still certain issues encountered by this research but has not been dug down sufficiently. They suggest further study for valuable policy implications such as: Impacts on the studying quality of the pupils who receive the support; Repetition of the study on the same group who will continue to receive support from Decree 49 so as to compare and collect policy implications on the scheme of implementation and levels of support. 63

76 Lessons learned for the future A number of lessons can be drawn from this study for future study using similar social audit approach. Lesson 1: Seek support and cooperation from local partners To ensure a study s success it is crucial to have the support and cooperation of the local authorities and people. Given the geographical remoteness and complicity of the area, access to villagers without the support of the local people is very difficult and time consuming, if not impossible. Lesson 2: Have an interpreter who knows the local language and Vietnamese Dien Bien is a province where ethnic minority groups make up the majority of the population, and very different languages are spoken. The problem of local languages should be taken seriously as many of the survey respondents cannot or do not want to communicate in Vietnamese. A good interpreter is essential for field surveys. Lesson 3: More time and resources are needed to conduct a real PETS Due to limited time and resources, the scope of this PETS was limited to only eight communes (out of 112 communes and wards) in two districts (out of nine districts and townships) in Dien Bien Province. The sample was not big enough to represent the whole picture of implementation of D.112 given the diversity in geographic distribution, the complicity and the ethnic structure of the province. Lesson 4: Select the right time to conduct the fieldwork and interviews Attention should be paid to selecting the right time to conduct fieldwork. For this PETS, the survey questionnaire pre-test and the field survey implementation were conducted in the January and February, just before Tet (Lunar New Year in Viet Nam). At this time, as expected, local government offices, schools and village people were all in a rush to prepare for Tet. This explained why there was very little willingness to fill out the survey questionnaire and be interviewed. More importantly, because of rain January and February was a very bad time to go to very remote villages. Very often the roads were unusable. Lesson 5: More time and effort are needed for discussion of the questionnaire, the pretest and finalisation Questions asked of the local people should be precise and easy for them to understand and answer correctly. This is not easy given that the survey target is ethnic people. More time should be given to have a proper, meaningful discussion and better elaboration of the questionnaire. Lesson 6: Technical support from international experts for report writing is needed PETS as a study technique is still new for many Vietnamese researchers. Therefore, it would be much better if instead of providing comments, an international consultant could spend more time working directly with the team at the stage of report writing and report elaboration. 64

77 References 1. Committee for Ethnic Minority Affaires (CEMA), Circular 06/2007/TT-UBDT 2. Dien Bien DOET, a report on the implementation of D.112 of 26 October Dien Bien PPC s Decision 105/QĐ-UBND of 5 February Ministry of Education and Training (MOET), Circular 24/2010/TT-BGDDT of 2 October 2010 Regulations on the Performance of CBSCs 5. Ministry of Finance (MOF), Circular 90/2011/TT-BTC guiding state budget estimations 6. MOF-MOET, Circular 109/2009/TTLT/BTC-BGDĐT 7. Prime Minister s Decision 85/2010/Decision-TTg 8. Project document of Program 135 ( ), phase III, 8th draft 9. Reports provided by government departments in Dien Bien to the PETS team during its inception visit in October Resolution 49/2010/NĐ - CP 11. The Prime Minister s Decision 112/2007- QD/TTg 12. UNDP (2008), a mid-term review of the implementation of Program 135-II 13. UNESCO, Truong Huyen Chi (2009), Schooling as Lived and Told: Contrasting impacts of education policies for ethnic minority children in Vietnam seen from Young Lives Surveys 14. UNICEF and Dien Bien PPC 2010, An analysis of the situation of children in Dien Bien, a report undertaken by the Provincial Child Friendly Program supported by UNICEF Viet Nam UNICEF, MOLISA (Social Protection Department & Institute of Labor Science and Social Affairs, 2010), an overview of the current policies regarding cash transfers for children within the social protection system in Viet Nam 16. UNICEF/IS/VASS (2008), a mid-term review on the implementation of Program 135, Component IV 65

78 Appendix Table A1: Main characteristics of the household sample Muong Cha District Dien Bien Dong District Number of responses Age Sex Ethnic group Education: Total Male % Female % Thai % Hmong % Other % Finished first grade or never went to school % Finished grade % Finished grade % Vocational training Marital status Single % Married % Widowed % Separated/Divorced %

79 Muong Cha District Dien Bien Dong District Total Average number of children 6-21 years old in the family Had a yellow card for the year % Had a yellow card for the year % The child was still going to school Average annual income (million VND) Types of assests available Kind of material the house was made of % Mean Median Bicycle % Motorcycle % Cell phone % Computer % Electric or gas stove % TV % Refrigerator/freezer % Other % Brick % Wood % Mud/straw % Other %

80 Muong Cha District Dien Bien Dong District Total Lighting Electric % Oil lamp % Wood/straw % Housing characteristics: Roof Sheet % Cement/concrete % Clay % Mats % Other % Housing characteristics: Floor Tile % Cement/concrete % Wood % Clay % Other % Source of water Private tap % Public tap % Public water tank % Surface water/streams

81 Muong Cha District Dien Bien Dong District Total % Well % Other % Type of toliet Flush % Turkish toilet % Latrine % Tham doi nuoc % Temparory latrine % No toilet % Other % Source of fuel Electricity/gas % Charcoal % Oil % Wood % Other % 69

82 Table A2: Main characteristics of the school sample Muong Cha District Dien Bien Dong District Total Number of responses Types of school Primary % 61% 59% 60% Lower secondary % 36% 41% 38% Both primary and secondary % 4% 0% 2% Number of schools with boarders % 53% 62% Average number of classrooms at each school Average number of built classrooms at each school Average number of classrooms with electricity at each school Number of days the power was cut in the last month None % One % Two to eight % About half the time 3 3 % 4.62 Most of the time % No power at all %

83 Muong Cha District Dien Bien Dong District Total 12. Separate toilets for males and females Yes % 93% 91% 92% No % 7% 9% 8% Access to a telephone Yes % 93% 95% 94% No % 7% 5% 6% Main type of telephone Landline % 62% 25% 40% Mobile phone (of the school) % 23% 31% 27% Mobile phone (of staff) % 15% 44% 32% 16. Newspaper subscription Yes % 74% 51% 61% No % 26% 49% 39% 17. Internet connection Yes % 79% 81% 80% No % 21% 19% 20% 71

84 Muong Cha District Dien Bien Dong District Total 18. Number of computers with Internet access for students to use Section 3. Time of travel by motorbike from school to (min) % Health facility or hospital Commune Centre District Treasury District DOET Budget Allocation Process Table A3: General budgeting process Stages Steps Time framework Budget preparation The Standing Committee of the National Assembly gave opinions on the budget allocation norms and the important expenditure regimes to provide the basis for budgeting. The Prime Minister issued a directive on formulating state budgets. The MOF and the MPI issued a circular to provide guidelines Ministries, central agencies and the Provincial People s Committee guided their subordinate units as they prepared budgets. The agencies, units and provinces sent their budgets to the MOF and the MPI. Before 1 May Before 31 May Before 10 June No later than 20 July 72

85 Stages Steps Time framework Budget appraisal, approval and allocation The MOF worked with the central and local agencies to consolidate and prepare the state budget and submit it to the Government. The Government submitted this budget to the National Assembly for appraisal The Committee on Budgetary and Economic Affairs appraised: (1) the execution of the state budget and the central budget for the year, and (2) the central budget allocation plan for the following year The Government prepared a justification report after the appraisal meeting of the National Assembly s agencies Appraisal report of the Committee on Budgetary and Economic Affairs of the National Assembly The Standing Committee of the National Assembly gave opinions. The Committee on Budgetary and Economic Affairs completed the appraisal report before the National Assembly session began. The National Assembly held a meeting to discuss, question and decide on the state budget and allocation of the central budget. The Government assigned budgets to the ministries, central agencies and provinces. From the end of July Before 1 Oct. No later than 5 Oct. Three days after the appraisal meeting No later than 12 Oct. Before 18 Oct. Within 10 days Before 15 Nov. Before 20 Nov. The provincial People s Committee reviewed and adjusted their budget on the basis of the decisions of the National Assembly, and then the provincial budget plan was submitted to the Provincial People s Council for discussion and decision. The Provincial People s Council made a decision on the provincial budget and the plan for its allocation. Before 10 Dec. The District People s Council decided the district s budget and the plan for allocating the district s budget. Budget execution Budget final accounts settlement Within 10 days of the date of approval by the Provincial People s Council Budget allocation notification to communes and budget spending units. The Government and the People s Committees at all levels organized the execution of the approved budget. The Government and People s Committees at all levels organized the preparation of the state budget final accounts and submitted them to the People s Councils at respective levels and the National Assembly for approval. Before 31 Dec. In the fiscal year 18 months from the final day of the fiscal year Source: Adapted from the State Budget Law

86 Amount of support received by households 30% of the students in the surveyed households (147 cases) reported having received the cash transfer support only once a year. Among them, around 71% (81 of the 115 beneficiaries) in Muong Cha District and 75% (24 of the 32 beneficiaries) in Dien Bien Dong District received the support between August and October of 2011 (see Table 4.17). This was most likely the support for the second semester of the school year, which was released on 22 June 2011 and 11 August 2011 in Dien Bien Dong District and Muong Cha District respectively. Table A4. Date of receipt by households that received only one payment for the school year Muong Cha District Percentage and number of observations Dien Bien Dong District November % (4) 0% (0) May % (1) 0% (0) September % (6) 3% (1) November % (4) 0% (0) December % (5) 0% (0) January % (8) 6% (2) March % (0) 6% (2) May % (1) 6% (2) June % (1) 0% (0) July % (1) 0% (0) August % (10) 0% (0) September % (44) 38% (12) October % (27) 28% (9) November % (1) 9% (3) December % (1) 3% (1) Total 100% (115) 100% (32) Source: From the household survey. Half of the recipients said they had received 700,000 VND in support for five months. 74

87 75

88 MINISTRY OF PLANNING AND INVESTMENT 81A Tran Quoc Toan, Ha Noi, Viet Nam Tel: (+84.4) Fax: (+84.4) Follow us: B Hoang DIeu, Ba Dinh, Ha Noi. Tel: (84-4) ; Fax: (84-4) Web: Ha Noi, January